(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Scientific papers of Asa Gray"

JDcTli 



Or 



SCIENTIFIC PAPERS 



OF 



ASA GRAY 



SELECTED BY 



CHARLES SPRAGUE SARGENT 



VOL. I. 

REVIEWS OF WORKS ON BOTANY AND 
RELATED SUBJECTS 

1834-1887 



37^^35=™^ 










^ffiffiSS^. 



BOSTON AND NEW YORK 

HOUGHTON, MIFFLIN AND COMPANY 

(©be fitoersi&e Prcsg, Cambri&ge 

1889 



Copyright, 1889, 
By CHARLES SPRAGUE SARGENT. 

All rights reserved. 



/tin \ir. 



^ 



The Riverside Press, Cambridge: 
Electrotyped and Printed by II. 0. Houghton & Co. 



INTRODUCTION. 



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 
place 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 ^ 



iv INTRODUCTION. 

a number of papers which owe their existence to the discus- 
sions which followed the publication of Mr. Darwin's " Origin 
of Species," — discussions iu 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 srreat mass of botanical students. The value of these 
papers, however, is historical only, as all that they contain of 
permanent usefulness has already been incorporated in stand- 
ard works upon the science, or will 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. Xo 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 upon 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. 

V*>. O- k5. 
Bkookline, 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 

Moqutn-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.e 87 

Kadlkofer's Fecundation in the Vegetable Kingdom . . 91 

Hooker on the Balanophoreje 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 

Cuktis'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 



Vlll CONTENTS. 

Naudin on Heredity and Variability in Plants . . . .212 
Darwin's Cross and Self- Fertilization in the Vegetable King- 
dom 217 

Phytogamy 241 

Bentham' s Flora of Australia 246 

Db Candolle's New Monographs 248 

Epping Forest 253 

Hooker and Ball's Tour in Marocco 255 

Bentham on Euphokbiacee 259 

Henslow on the Self-Fertilization of Plants .... 263 

Plant Archeology 269 

Watson on North American Liliacee 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 Peruvian Andes ...... 384 



REVIEWS. 



LINDLEY'S NATURAL SYSTEM OF BOTANY. 1 

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 Richard'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 

1 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. London : 1836. (Ameri- 
can Journal of Science and Arts, xxxii. 292.) 



Z 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 produce 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 Linnaeus." 1 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 in 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- 
biferce also in the Linn?ean class Syngenesia. I see no good reason for 
these innovations." Now Linnaeus, in his artificial arrangement, certainly 
did place these genera (and also Parthenium and Iva) in Monoecia Pen- 
tandria ; but the Innovator in this instance is Jussieu himself, who never 
referred these two genera to Urticece, but places them in his order Corym- 
biferoz (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. o 

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 country, 
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 opinion 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 live 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, properties, 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- 
sjiermce, or Flowering plants with naked ovules (comprising 
the Coniferce, Cycadea?, and, according to Brongniart and 
Lindley, the Equisetacece), and the Rhimnthce, 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, Mushrooms, 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, w r hile 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 onh r , 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 



b HE VIEWS. 

processes ; Endogens, having no occasion for such 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 Exorhizcv, and the latter Endo- 
rhizoe. 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 synonymous 
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 cotyledons than two, whence 
they have been called Polycotyledons ; their radicle usually 
adheres to the albumen in which the embryo lies, and that 
circumstance has given rise to the name Synorhizm. The 
veins of their leaves, when they have any veins, are either 



LINDLEY'S NATURAL SYSTEM OF BOTANY. 7 

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 Rkizanthm, 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 apparent 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 lanre 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 Eobert 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 OE BOTANY. 9 

now employed was first sketched in the " Nixus Plantaruni " 
(1832), and afterwards in the " Key to Structural, Physio- 
logical, and Systematic Botany" 1 (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 Acrogeus, and Rhizanthce, which form the transi- 
tion from Endogens to Acrogens of the lowest grade. The 
great class Exogence (Dicotyledones of Jussieu) is divided 
into three subclasses, namely : — 

1. Pohjpctalas ; those with the floral envelopes consisting 
of both calyx and corolla, the latter composed of distinct 
petals. 

2. Mbnopetalce ; those with the petals combined in a mono- 
petalous corolla. 

3. Jncompletce ; 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 polypctalous orders into seven, and the Monopetaloe, and 
Incompletes 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. Epigyinosoe. Ovary inferior, usually having an epigy- 
nous disk. 

3. Parietosce. Placentation parietal. 

4. Ca/ycoscc. Calyx incompletely whorled ; two of the 
sepals being exterior. 

5. Syncarposce. None of the characters of the other 
groups, and with the carpels compactly united. 

1 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. 

G. Gynobaseosce. Carpels not exceeding five, diverging 
at the base, arranged in a single row around an elevated axis 
or gynobasc. Stamens usually separate from the calyx. 

7. Apocarposce. None of the characters of the other 
groups, but with the carpels distinct, or separable by their 
faces, or solitai'y. 

Next, every group is divided into smaller groups, each of 
which includes one, two, or several orders. These minor groups 
are called Alliance*, and are distinguished by the termination 
ales. Thus, under the Albuminose group, we have : — 

Alliance 1. Handles, comprising the Ranunculacece, Pa- 
paveracem (with its suborder, as Lindley, following Bern- 
hardi, considers it, Fumariece), Nympliaiaccaz, (to which 
Hydropeltidem is improperly joined), and Nelumbiaceai ; 

Alliance 2. Anonales, which comprehends the Nutmeg 
tribe, the Anonacece, Magnollaceai, etc. ; 

Alliance 3. Umbellales, including the Umbelliferous tribe, 
with the nearly allied Araliaccaz ; 

Alliance 4. Grossales, consisting chiefly of the Grossu- 
lacece or currant tribe ; and lastly, 

Alliance 5. Pittosjwrales, which strikes us as a singularly 
heterogeneous assemblage, bringing together into one group 
the Vitacece, Pittosporacece, Olacacece, Francoacece, and Sav- 
raceniaceai. 

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, Labiatce, Cmcifcrce, Umbelliferai, 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 Jft/rti, Lilia, Trides, Euphorbice), 
or with a slight prolongation (as Orchidece, Jasminece, etc.), 
or with the termination still further modified (as in Cyper- 
oidece, Aroidea\ Boragineoe, or Ranvncirfacea?, Rosacem, 
Cuciirbitacece, 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 acece 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 xise which relate to some peculiarity in the habit 
of the family. Thus instead of Criicifercp, UmbelUferm^ 



12 REVIEWS. 

Conifer w, etc., we have Cruciacece, Umbellacece, 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. 
Brassicaccce, Ajriacece, and JPinacece 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 
Vegetabilium," which was finished in 1827, contains (exclu- 
sive of the appendix) only 3593 genera, or not quite half the 
number now known ; while the twelfth edition of " Systema 
Naturae " (the last of Linnaeus 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 
granted 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 
Linnaeus 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 with 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. 

" Caesalpinus, Adanson, Jussieu, and Gsertner, 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, which 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 Ra- 
nales, which stands at the commencement of Dr. Lindley's 
treatise. AVe 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, Nymphceacece, etc.), or points in the opposite di- 
rection (as in Neliimhiaceve and Cabombacece) ; 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 " 1 — with which our notices may appropriately 
commence — was published at the very close of last year, and 
comprises the following orders, namely, Stylidece, Lobeliacece, 
Campamdacece, Cyphiacece (a very small order, founded on 
the Cape genus Cyphia, and here first proposed by Alphonse 
De Candolle), Goodenoviece, Gesneriacece, Napoleoneoe, Vac- 
cinece, Ericaceae, Epacrideai, Pyrolacece, Francoacece, and 
Monotropem. Of these, the Lobeliacem, Campanulacece, and 
CypMaccce were elaborated by Professor Alphonse De Can- 
dolle, the well-known son of the distinguished author ; the Vac- 
cines 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 Encacece nearly 
in the manner first proposed in the " Theorie Elementaire," 
considering the Vacdniece, Monotropem, 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 Xuttall (which was founded upon 
Cyrilla paniculata 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 Vegetabilis, 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 XIIL, 
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, 1S70 
(Ibid., 2 ser., xlvii. 125). — Pars XVII. etc. Paris, 1873 (Ibid., 3 ser., 
vii. 66). 



16 HE VIEWS. 

cated to Dr. Pickering a curious Leguminous plant from Cali- 
fornia. The genus Galax, De Candolle has appended to Py- 
rolacece (tribe Galacece), 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 galadfolia. 
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 
Compositoe (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 Compositor 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 : — 



BE CANDOLLE' S PRODROMUS. 17 

1. TuBULiFLORiE ; those with the perfect flowers tubular 
and regularly five (or rarely four) toothed. 

2. Labiatifloile ; those with bilabiate, or two-cleft, per- 
fect flowers. 

3. LiGULiFLORiE ; w r hich 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, Vemoniacea . 
Eupatoriacece, Asteroidece, Senecionidece and Cynarece. The 
second series consists exclusively of the Jfutisiacece and the 
JVassauviacece, chiefly South American plants ; a single spe- 
cies of Chaptalia is, w r e believe, the only North American 
representative. The third series, comprising the Chicoracea , 
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 PlumbagineoB ; Mr. Duby of Geneva will prepare the 
Primulacem ; Professor Dunal of Montpelier, the Solanece ; 
Mr. Decaisne of Paris, the Asclepiadece ; and Mr. Bentbam, 
the Scrojihularinece and Babiatce. 

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, 1 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 Boganiacece, Bignoniacece, 
Cyrtandracece, Sesamece, and Borraginem, from the notes 
prepared by the late Professor De Candolle ; the Hydrophyl- 
lacece, by Alphonse De Candolle ; the Gentia?iacece, by Grise- 

1 The elder De Candolle died September 9, 1841. 



18 REVIEWS. 

bach; the Polcmoniacccv, by Bentham, and the Convolvula- 
cece, by Choisy. The tenth volume will be occupied with the 
Solanacece, by Dunal, and the Scrophularinece, 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 Lcntibulariece, 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. resupinata, discovered by 
B. D. Greene, Esq., and first mentioned in the Massachusetts 
Catalogue of Plants and Animals, 1835, is wrongly placed 
by De Candolle 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. Naumburgia thyrsijlora, 
Mcench, = Lysimachia thyrsiffora, Linn., and L. capitata, 
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. florihundas, HBK. S. ebracteatus is not 
noticed as a plant of the United States, although it is com- 
mon along our southern borders, nor is it distinguished 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 31yrsi?iacece, 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 PicJceringia, 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 Styracece 
(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 Oleacece is published from the manuscripts of 
the late Professor De Caudolle. The American species of 
Fraxinus still require the labors of a monographer. 

The order Jasminece is made to comprehend Bolivaria (of 
which there is at least one Texan species) and Menodora ; 
and the family Bolivariacece is shown to have been founded 
upon misconceived characters. 

For the elaboration of the Apocynacece, 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 (F. dif- 
formis, DC, = Echites difforrnis, 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. tilicefolius'), and 
one, or possibly two species of Chthamalia, Decaisne. Me- 
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 Selectae " 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, 
especially 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 Ion o-- 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 Bignoniacem is edited from the manuscripts of the 
elder De Candolle ; as are also the orders Sesamem and Cyr- 
tandracecB, which last has been reduced by Mr. Brown to Ges- 
neriacece. The Hydrophyllacece 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, Dactylophylluin, 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 
" infelicissiiue intricatus " is perhaps still applicable to their 
family ; and the genera are probably unduly increased in 
number. Of the Borraginece, printed from the elder De 
Candolle's manuscripts, with valuable notes and additions by 
the editor, we have the first three tribes, namefy, Cordiece, 
JShretiece, and JBeliotropece. 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 Mas 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 Orobanchacem, by Mr. 
Eeuter of Geneva ; the true genera of which are Epiphegus 
of the United States ; Phelipaea, containing thirty-eight species 
of the Old "World, and two of North America west of the 
Mississippi; Conopholis (for the Orobanche Americana, L.) ; 
Orobanche of over ninety species belonging to the temperate 
regions of the Old World : Boschniakia of two high north- 
west American and Siberian species ; Clandestina of one south 
European species ; Lathrsea 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 species 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 
uni flora 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 imposed, was 
adopted in the MSS. for this family prepared for the " Flora 
of North America," and also in the writer's " Manual of the 
Botany of the Northern United States." Besides A. uni- 
Jlonan, we have in the West and North, A. comosum, and A. 
fasciculatum. 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 Aeanthacece 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 Ruellia 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 Ruellia? belong to Dipteracanthus 
and Calophanes. Our Diantliera Americana, L., with its 
allies, is included in the large genus Rhytiglossa, 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, we 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 Justicla 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 Verbenacece 
as well as Labiatce, 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 Verbenacece, composed of three tribes (the Ver- 
benece,Viticece,Avice?i?iiece), 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. 

AVe 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 Labiates 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 Jfonopetahr, with the ex- 
ception of the small family Plantaginacece, 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 Labiatce, 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 
Menihoidece and the Melissinece of the earlier work, and the 
Seidell arinece 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 /Satureiece we have Mentha, Lycopus, a single Cu- 



24 REVIEWS. 

nila, Pycnanthcmum (the whole seventeen species), an anom- 
alous Satureia (S. rigida, Bartr.), three species of Microme- 
ria ; while to Calamintha is now referred the former M. gla- 
bella and M. Nuttallii, as well as the Gardoquia ffookeri, 
Benth., with the new C. canescens, Torr and Gr., MSS., and 
C. Caroliniana, Sweet (Thymus, Michx.) ; also two species 
of Dicerandra (of which D. dcnsiflora is a new one from 
Florida) ; the California Pogogyne ; Hedeoma, including 
//. dliata (Keiihia ciliata, Benth., Lab.) ; and Collinsonia, 
of which six species are recognized. 

Of the tribe Monardeai 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 M. fistulosa 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 Stachydece 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, Engelin. — Gray (in which 
the synonym " Physostegia truncata, Hook. Botanical Maga- 
zine, t. 3494," should be cited under B. seutellarioides, 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 Bras ice belongs entirely to the Old World and to 
the Sandwich Islands ; and the tribe Prostanthereaz is exclu- 
sively Australian. 

Of the Ajugecs 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 Gymnandra is 
doubtfully appended. Of three Siberian species of this genus 



DE CANDOLLE'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 Stilbacecc, 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 species, and a new one of a single 
species ; all of Europe and of Eastern Asia, except one in the 
Canary Islands. 

The order Brunoniacece, 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 JPlumbaginacece. The 
tribe Staticece comprises six genera, namely, iEgialitis, 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. Carol i?iia?ia, Walt., Mr. Boissier distinguishes from 
S. 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 Phimbagece 
consists of the Siberian Plumbagella, the European and tropi- 
cal Plumbago, the Abyssinian Valoradia, and the African and 
North Indian Vosrelia. 

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 Solanacem and the 
JPlantaginacece, two families which will finish the Monopet- 
alous series, as this begins the Apetahu or Monochlamydece. 
It comprises the Phytolaccacece, Salsolaceai (Chenopodeai), 
Basellacecv, and Amarantacece, elaborated by Moquin-Tandon 
of Toulouse, and the Nyctaginacem, by Professor Choisy of 
Geneva. Of JPhytolaceacece we have in the United States, 
only Petivera alliacca, which grows in Florida (probably not 
in " Carolina"), liivma Icevis (to which we are surprised to 
see M. portulaco'ides, 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 Cyclolobeai (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- 
podium Virginicum to C. aristatum, but it is doubtful if we 
possess the genus ; Cycloloma (Salsola platyphylla, Michx.) ; 
Chenopodium, to which Tandon now reunites the greater part 
of his Ambrina (C. ambrosiodes, C. < in t hclminticum, etc.), 
leaving in Roubieva only the original species, recently illus- 
trated in this Journal by Mr. Carey ; Blitum, to which the 
author now refers, as a section, his former genus, Agathophy- 
ton (Chenopodium Bonus Henricus,Jj.^) ; Monolepis, Schrad. 
(Blitum chenopod hides, Nutt.) ; Atriplex, of which too many 
of the older species are credited to the United States ; Obione, 
Gsertn., of which nine species are North American, including 
(apparently with sufficient reason) the Pterochiton, Torr., 
Grayia, Hook, and Arn., 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 ? S. Peruviana (Caro. Eraser), and S. 



DE 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- 
throcnemum (A.? ambiguum — Salicomia ambigua, Michx.) 
being still kept distinct. Of the Spirolobece we have in 
North America, Chenopodina, a genus newly founded for the 
Chenopodium maritimum, L., which was formerly referred 
to Suasda, besides which species Tandon also gives us C line- 
aris, Ell., 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. calceoliformis ( Chenopodium calceoli- 
forme, Hook.), which is stated also to be found " near New 
York " ; of Salsola, we have S. kali 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 fiasellacece, familiar to us only by the Bous- 
singaultia baselloidcs, which is cultivated as an ornamen- 
tal climbing plant, contains six genera, entirely of tropical 
plants. 

The order Amarantacece 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 Blitum ; 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 tamariscinus, Nutt., which 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 (Halomocne- 
mis occidentalism Nutt., ined.) ; an obscure Polycnemum ; 



23 REVIEWS. 

Gossypianthus, Hook., two Texan species ; Iresine, two spe- 
cies ; Alternauthera, one species (Achyranthus i'cj)e?is, L.) ; 
besides the A. (Cladothrix, 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 ficoidca and T. polygonoides 
appear to be only introduced plants along our southern coast. 
Frselichia (Oplotheca, Nutt.) has three North American 
species. Phyllepidium of Rafinesque is not identified and 
probably never will be. 

The remaining family, JSfyctaginacece, includes eighteen 
genera, in three tribes. Of Mirablis, though no species 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 undescribed Texan representatives of the family. 

The second part of the fourteenth volume of the " Prodro- 
mus " contains the T7iymelmacea2 by Meisner, the Elceagnacece 
by von Schlechtendal, the Grubbiaccaz 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 Santalacem 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 Elceagnacece. As to our few Santalacem, it is interesting 
to remark that one of our characteristic genera, Pyrularia 
(the Oilnut), is found to have two representatives in the 
Himalayas (Sphaerocarya, Wall.), and apparently two more 
in southern India (Scleropyrum, Am.). 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 ( O. pallida) 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 Loranthacecc and 
Protcacem also. Our author's views are presented in detail in 
an article, " Sur la Famille des Santalaceae," in the " Biblio- 
theque Universelle," published last autumn, and they appear 
wellnisrh convincing:. An analogous case is found in Zan- 
thoxylum (only here the suppression is the rare case), Z. 
Americanum plainly wanting that which in Z. Oarolinianum 
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- 
ledonece) 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 Eauracece 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 Vriese 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 apjiended, and perhaps the Euphorbiacece, also by De 
Candolle, except the genus Euphorbia which Boissier under- 
takes. The sixteenth volume is intended to commence with 
the Lrticaccce proper, by Weddell, or the Jfonimiacece by 
Tulasne. We are pleased to learn that Professor Anderson 
of Stockholm is to elaborate the Salicinece. 

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 Euphorbiacece, by Dr. Miiller ; the Empe- 
tracece by Alphonse De Candolle himself (Empetrum reduced 
to one species, Corema of two, and a Ceratiola) ; Cannabinem 
by the same (the Ulmacece and Artocarpece postponed not 
being ready), the Urticacece (i. e. the Urticece) by Weddell ; 
Piperacece by Casimir De Candolle (the Saururece made a 
mere tribe, and the Pipereai mainly included under Piper of 



DE CANDOLLE'S PRODROMUS. 31 

635 species, Peperomia of 389 !) ; Chloranthacecs by Solms- 
Laubach of Halle, reduced 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. buxifolia, 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, Sarraceniacece, PhytocrenecB, Car- 
diopteridece, Salvadoracece, Cynocrambe, Batidacece (one Ba- 
tis), Lennoaceai (by Solms-Laubach), Bodostemacece (by 
Weddell). Then Nepenthacecc, by Dr. Hooker; Cytinacea', 
by the same ; Balanophoracece, by Eichler ; Ulmacece, by 
Planchon ; Mbracece, by Bureau ; and a synopsis of the genera 
of Artocarpece. For the complete elaboration of the last 
famiby 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 Artocarpece 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, Kumeri, 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 



32 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 Dicotyledon es. Among a few statistical 
data which are given, the ten orders arc enumerated which 
contain the greatest number of genera, beginning with Com- 
posites and ending with Cruciferce. Then the ten which 
most abound in species, which begin with Composite (8561 
species) and end with Umbelliferce (1016). Leguminosece 
are the second in both lists, and next Rubiacece in the former 
and Eupltorbiacece in the latter. But the long interval be- 
tween the publication of many orders, say between Cruciferce 
and EupJiorbiacctv, 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, elaboi*ate 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 sur les Ouvrages Generaux cle Bota- 
nique Descriptive." 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 
Vegretabilium " 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. 



ENDLICHERS GENERA PLANTARUM. 

This is one of the most important works 1 of the age ; 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 the 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 Endlicher's Genera Plantarum secundum Ordines Naturales disposita. 
Vienna, 1836-40. (American Journal of Science and Arts, xxxLx. 17G.) 



34 REVIEWS. 

this time appeared. It is 3tated in the original announcement 
that the work will not exceed ten or twelve numbers ; we 
imagine, however, that four or live 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 fust edition of the " Genera Plantarum " by Linnaeus 
was published at Leyden in the year 1737 ; the second and 
third were 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 Linnaeus 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 Linnaeus is, we 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 
unelassical names of Aublet and others for new ones made 
according to the Linnaean canons. Succeeding authors in 
plucking these borrowed plumes have despoiled him of some 
rightful 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 genera is here increased to 1769. About the same time 
(1791) an edition was published by Haenke at Vienna, which 
is apparently carefully digested. The latest edition of the 
" Genera Plantarum " which bears the name of Linnaeus, and is 
arranged according to the artificial system, is that of Sprengel, 
and published at Gottingen in 1830 and 1831 (2 vols. 8vo), 



ENDLICHER' S GENERA PLANTARUM. 35 

which is the latest complete work in which the known genera 
are characterized. He gives the date of the publication of 
each genus, and references to the principal figures. The whole 
number of the genera described is 4159. 

The " Genera Plantarum secundum Ordines Naturales dis- 
posita " of the immortal Jussieu, with which a new era in bot- 
any commenced, appeared in the year 1789. This work has 
never been reprinted in France, and but once out of it, and is 
now very scarce. Until the commencement of Dr. Endlicher's 
work, a period of about half a century, it has remained the 
only Genera Plantarum according to the natural system. 
There is but one living botanist upon whom the task of pre- 
paring a new Genera of Plants would seem most appropriately 
to devolve ; but since it cannot be expected from that quar- 
ter, we are glad it has been undertaken, and we may almost 
say completed, by so learned and careful a botanist as Dr. 
Endlicher. The onlv fault we have to notice is, that there 
is no mode of distinguishing directly the generic characters, 
which are compiled altogether from preceding authors, from 
those drawn from the plants themselves. An author can only 
be considered responsible for the latter ; yet unless there be 
some means of distinguishing those which have been verified 
from the remainder, he becomes somewhat implicated in the 
mistakes of his predecessors. Dr. Endlicher being scarcely 
less distinguished as a classical scholar than as a botanist, this 
work is a perfect model of the classical style. 

Simultaneously with this work, which it is in part intended 
to illustrate, the author is publishing an " Iconographia Ge- 
nerum Plantarum." It appears in quarto parts, with about 
twelve uncolored plates in each, executed in a very superior 
manner, with full analyses, which leave nothing to be desired 
in this respect. Seven or eight parts are already published. 
It is the cheapest illustrated work of the kind with which we 
are acquainted, and at the same time one of the very best. 



36 REVIEWS. 



HARVEY'S SOUTH AFRICAN PLANTS. 

Tnis volume 1 was written, printed in very handsome style, 
and published at the Cape of Good Hope. It was prepared, 
not, as we might suppose, for the purpose of making Cape 
plants better known to European botanists, but for the use of 
the students and lovers of flowers at the Cape. It is arranged, 
moreover, according to the Natural System, and is throughout 
a work of genuine science. Truly, if popular botanical works, 
based on the Natural System, are deemed most advantageous 
for students at the Cape of Good Hope, we may indulge the 
expectation that this method will in due time be universally 
adopted in Europe and the United States. Mr. Harvey, who, 
while occupied with his duties as colonial secretary, has been 
enabled to do so much for the botany of that rich and inter- 
esting region, both by his own researches and by encouraging 
the labors of others, was requested to recommend some intro- 
ductory work on botany. 

Had a mere introduction to the elements of the science alone 
been needed, the desideratum might easily have been supplied. 
" But I soon found," says Mr. Harvey, " on cross-questioning, 
that something very different was required. One lady told 
me that she knew already what 'calyx, corolla, stamens, and 
pistils, and all that ' meant ; and another had penetrated the 
mystery of Monandria, Diandria, etc., and did not want to be 
told that over again ; what they desired was a book in which 
they could discover the names of every plant that struck their 
fancy in rambling through the fields, — in short, a Flora 
Capensis. Here I found myself completely at fault, for there 
seemed little use in recommending the Flora of Thunber<r, or 
the more ancient writings of Burmann ; for even could they 
be procured, which would not be without much difficulty, they 
would have proved perfectly useless to my lady friends, who, 
not being blue-stockings, could have derived little instruction 

1 The Genera of South African Plants, arranged according to the nat- 
ural system, by William Henry Harvey, Esq. Cape Town, 1838. (Ameri- 
can Journal of Science and Arts, xxxix. 173.) 



FLORA OF JAPAN. 37 

from the crabbed Latin in which they were written." Mr. 
Harvey then conceived the idea of writing a Flora Capensis ; 
but it at once occurring that such a work must consume a 
long series of years in preparation, he decided upon rendering 
that more prompt, though less complete assistance, which a 
work like the present is calculated to afford. " The Genera 
of South African Plants " is the result of this determination ; 
for which the author deserves the thanks, not only of the lady 
friends whose benefit lie had chiefly in view, but of all the 
cultivators of botanical science. Although much more time 
would be required for its preparation, the work would have 
been more valuable had Mr. Harvey placed still less de- 
pendence on preceding authors, and drawn his characters, in 
every practicable instance, from the plants themselves ; but 
only those who are accustomed to prepare their works in this 
manner are aware of the vast amount of labor it involves. 
The general plan of the work, as the author informs us, is 
taken from Beck ? s " Botany of the Northern and Middle 
States of North America," and Nuttall's " Genera of North 
American Plants " ; in the arrangement and characters of the 
orders, Dr. Arnott has chiefly been followed. The number 
of genera described is 1086, distributed under 135 orders. 
Many South African genera have been published in still more 
recent general works or particular memoirs, or in those which 
have not reached the Cape in time to be employed by Mr. 
Harvey, so that the number of Cape genera may be safely 
estimated at 1200. 



SIEBOLD'S FLORA OF JAPAN. 

This work 1 is, we believe, wholly arranged and prepared 
by Professor Zuccarini of Munich, from notes and specimens 
furnished by Dr. Siebold of Leyden, accumulated during his 

1 Flora Japonica. Sectio prima, Plantse ornatui vel usui inservientes, 
digessit Dr. J. G. Zuccarini. Fasc. 1-10. Leyden, 1835-39. (American 
Journal of Science and Arts, xxxix. 175.) 



38 REVIEWS. 

long official residence in Japan. The admirable plates are 
executed in Munich : they are engraved upon stone after a 
peculiar method, which is now frequently employed, and are 
certainly not excelled in beauty or accuracy by any copper- 
plate engraving in the same style. The portion already pub- 
lished comprises only the ornamental or otherwise interesting 
plants, the general account of the Japanese flora being re- 
served for a future part of the work. The flora of Japan pre- 
sents such striking analogies to that of the temperate part of 
North America as to render this work of more than ordinary 
interest to American botanists. To show this, we select from 
the forty-six species described and figured by Zuccarini, the 
following list, placing opposite the Japanese plant the related 
North American forms. 

FLORA OF JAPAN. FLORA OF NORTH AMERICA. 

Illicium religiosum, Micium Floridanum and parvifiorum. 

Kadsura Japonica, Schizandra coccinea. 

Benthamia Japonica, Cornus florida. 

Corylopsis, two species, Hamamelis and Frothergilla. 

Aralia edulis, Aralia racemosa. 

Symplocos lucida, Hopea tinctoria. 

Styrax Japonicum, etc. Styrax, several species. 

Deutzia, three species, Philadelphia. 

Schizophargma Hydrangeoides, ) j Decumaria and 

Platycrater arguta, \ \ Hydrangea. 

Diervilla, several species, Diervilla Tournefortii. 

Viburnum tomentosum, Viburnum lantanoides. 

Wisteria (or, as it should be, ) 

Wistaria) Japonica, and two >- Wistaria frutescens. 

other species, ) 

Pauloionia imperialis, Catalpa cordifolia. 

While about half the species thus far published are nearly 
related to (chiefly characteristic) North American plants, only 
eight, besides those given above, belong to genera which have 
no representatives in this country. The list might be greatly 
extended by comparisons from other sources. Thus Hoteia 
Japonica of Morren and Decaisne (which belongs to the ear- 
lier established Astilbe, Don.), which was by Thunberg mis- 
taken for Spiraea Anmcus, closely resembles our own Astilbe 
decandra, which has been more than once confounded with 



FLORA OF JAPAN. 39 

Spircea Amncus. On some future occasion we hope to make 
a somewhat extended comparison between the flora of tem- 
perate North America, and that of Japan and Middle Asia. 1 

Professor Zuccarini, the author, in conjunction with Dr. Sie- 
bold, of the excellent " Flora Japonica " now in progress, has 
recently published the first part of a brief memoir, entitled, 
" Florae Japonic® familiae Naturales, adjectis generum et spe- 
cierum exemplis selectis : Sect. i. Plantae dicotyledonese poly- 
petalae." 2 It is interesting- to remark how many of our char- 
acteristic genera are reproduced in Japan, not to speak of 
striking analogous forms. Thus the flora of Japan has not 
only "Wistaria, Lespedeza, Sieversia, Chimonanthus (in place 
of our Calycanthus), Philadelphus, several species of Rhus 
closely resembling our own, and two peculiar genera of Ju- 
glandece, but also a Pachysandra, some Berchemias, a Sta- 
phylea, and a peculiar genus of the tribe (Euscaphis) besides ; 
not only a dozen Maples, but also a Negundo, a Stuartia, two 
Tilias, a Phytolacca, an Opuntia (surely not indigenous), a 
Sicyos referred to our own >$'. angulata, two Droseras, a 
Nelumbium, a Nuphar, and two species of Nymphaea, Gynan- 
dropsis, a real Dicentra (Dielytra) and an allied new genus, 
with several species of Corydalis, a Trollius, our own Coptis 
and two new ones like the western C. asplcnifolia, an Iso- 
pyrum, two species of Aquilegia, one of them near A. Can- 
adensis, a Cimicifuga, a Trautvetteria, an Illicium, some 
Magnolias, Kadsura and Sphserostemma in place of Schi- 
zandra, a Mitellopsis, two species of Astilbe (Hoteia), many 
Hydrangeas as well as peculiar Hydrangeaceous forms, a 
Hamamelis with two other characteristic genera of the family, 
some true Dogwoods, as well as Benthamia the analogue of 
our Cornus Jiorida, some true Vines, and two species of Am- 

1 This short paper is of peculiar interest. It contains the earliest record 
of Professor Gray's investigations into the flora of Japan, and its relations 
to that of eastern North America — investigations which many years later 
enabled him to explain the distribution of plants through the northern 
hemisphere by tracing their direct descent from ancestors wh'ch flour- 
ished in the Arctic region during the latest tertiary periods, and estab- 
lished his reputation as a philosophical naturalist. — C. S. S. 

2 American Journal of Science and Arts, 2 ser., ii. 135. 



40 REVIEWS. 

pclopsis, three species of Panax, and four of Aralia, one of 
which is near our A. midicaulis ; and among Umbelliferae 
are Ilydrocotyle, Sanicula, Siuin, Angelica, but what is most 
remarkable, Cryptotamia, Archemora, and Osmorhiza. Fur- 
ther cases of generic conformity abound in the remaining di- 
visions of the vegetable kingdom; thus, for example, Dier- 
villa, Mitchella, Madura, Liquidambar, Torreya, and Sassa- 
fras, are represented in the flora of Japan. 



VEGETABLE MONSTROSITIES. 

Tnis interesting treatise * on Vegetable Monstrosities is 
very properly prefaced by a statement of what is meant by 
the normal structure of plants, by vegetable individuality, 
and vegetable symmetry. The author proceeds to consider, 
first, those slighter deviations which are called varieties ; 
and secondly, those more grave and mostly congenital anom- 
alies which bear the name of monsters. As to the latter the 
author remarks, that nearly every monstrous or abnormal con- 
dition that has been observed is to be met with as the normal 
state of other vegetables ; and that between a monstrous and 
a normal flower, the only difference often is, that the former 
is the occasional, and the latter the habitual state. " La 
monstruosite est done, en general, l'application insolite, a un 
individu ou a, un appareil, de la structure norm ale d'un autre 
appareil ou d'un autre individu. (Test un organisation trans- 
posed, e'est une loi changee de place. On Fa dit avec raison, 
la monstruosite ne se trouve pas en dehors de la nature, mais 
seulement en dehors de la coutume." It is clear, therefore, 
that while abnormal states may always be explained by the 
laws which regulate the normal structure, monsters themselves, 
as the etymology of the name indicates, often show us the true 
structure when it could not be certainly inferred from the 

1 Ele.mens de Teratologic Ve'ge'tale, ou Hittoire abrege'e des anomalies de 
V Organisation dans les Vegetaux. Par A. Moquin-Tandon. Paris, 1S41. 
(American Journal of Science and Arts, xli. 374.) 



AGASSIZ'S ZOOLOGICAL NOMENCLATOR. 41 

habitual condition. The author arranges monstrosities under 
four primary classes : those of volume, of form, of disposition, 
and of number. These are divided, the first class into mons- 
ters by diminution of volume (Atrophy), and by augmenta- 
tion (Hypertrophy) ; the second class into monsters by alter- 
ation of form, whether irregular (Difformation) or regular 
(Pelorias), and monsters by the transformation of one organ 
into another (Metamorphosis) ; the third class into monsters 
by the abnormal connection of parts, or by the disunion of 
parts habitually united, and into those caused by change of 
situation, or displacement ; the fourth class into monsters by 
diminution of number, or abortion, and those by augmentation 
of number. Under these heads the monstrosities of the dif- 
ferent organs of plants are considered in detail, and in a phil- 
osophical and very interesting manner. This brief notice 
of the plan of Moquin-Tandon's work, we are confident, will 
suffice to commend it to the attention of the botanists of this 
country. 



AGASSIZ'S ZOOLOGICAL NOMENCLATOR. 

This great work, 1 which must have cost an extraordinary 
amount of labor, is now almost completed. Trusting that some 
one of our able zoologists will duly give an account of a work 
which is indispensable to every votary of their science, we pro- 
pose at this time merely to call attention to the preface, pub- 
lished last year with the ninth and tenth fasciculi, and to ex- 
press unqualified admiration of the manner in which a subject 
of interest to all naturalists, that of nomenclature, is there 
treated. While botanists are enjoying the benefits of a sedu- 
lous adherence to the wholesome rules imposed by the father of 
natural-history nomenclature, and of nearly unanimous agree- 
ment in the few changes which the progress of science and the 
multiplication of its objects have rendered needful, the zoolo- 

1 Nomenclator Zoologicus, cnntinens Nomina Systematica Generum Ani- 
malium, tarn viventium quamfossilium, etc. Auctore, L. Agassiz. Soleure, 
1842-46. (American Journal of Science and Arts, 2 ser., iii. 302.) 



42 REVIEWS. 

gists on the other hand, who have too generally allowed every 
one to do that which was right in his own eyes, are reaping 
in consequence a plentiful harvest of confusion. The dif- 
ficulty of a reform increases with its necessity. It is much 
easier to state the evils than to relieve them ; and the well- 
meant endeavors that have recently been made to this end are 
some of them likely, if adopted, to make " confusion worse 
confounded." Probablv no living zoologist is so conversant 
as Professor Agassiz with the actual state of the nomenclature 
of the animal kingdom, and so well qualified to judge of the 
practical working of proposed rules, which often involve con- 
sequences that the propounders never dreamed of. Our 
author's views are therefore entitled to great weight. We are 
glad to perceive that they entirely concur with those quite 
unanimously adopted in the other great department of natural 
history for which the Linnaean canons were originally framed. 
As these canons were the foundations of our nomenclature, 
Professor Agassiz has very propeily reproduced them, totidem 
verbis, from the " Philosophia Botanica," adding now and 
then a short but pithy commentary. He then proceeds to 
examine the rules proposed by the Committee of the British 
Association, and shows that while some of them are mere iter- 
ations of the Linnaean canons, which should never have lost 
their authority, others are contrary to them, or threaten greater 
evils than they are intended to remedy. In most respects his 
criticisms concur with those already made by Dr. Gould in a 
former volume of this Journal (XIV. p. 1). We agree with 
Professor Agassiz in thinking these English canons worthy of 
adoption only when they agree with the letter or spirit of the 
Linnaean rules, which indeed they generally do. Those which 
conflict with them have not received, and probably will not 
receive, the general assent even of British naturalists. Hence, 
in our opinion, the American Geological Association has too 
hastily reaffirmed them, while they have, indeed, improved 
their form in several respects. It may be well to notice the 
comments of Professor Agassiz upon the more objectionable 
propositions. 

Their first rule, " that the name given by the founder of a 



AGASSIZ' S ZOOLOGICAL NOMENCLATOR. 43 

group or the (first) describer of a species should be perma- 
nently retained," cannot be too firmly insisted on ; for upon 
it rests the stability which is the most essential requisite of 
nomenclature. Their second rule, that since "the binomial 
nomenclature originated with Linnaeus, the law of priority is 
not to extend to the writings of antecedent authors," restricts 
the former too arbitrarily, and conflicts, as Professor Agassiz 
states, both with the canons and the example of Linnaeus, not 
less than with the conscientious practice of good naturalists 
ever since. Linnaeus was not the founder of genera or of 
generic nomenclature, and, " far from making new names in 
every instance, he retained all names given by his predeces- 
sors, provided they could be received into his system." It is 
generally thought that Linnaeus erred by adopting, not too 
many, but too few of the unobjectionable and well-established 
generic names of his predecessors, such as Tournefort, etc. 
Now when, in the natural progress of the science, a Linnaean 
genus is resolved into two or more Tournefortian ones, for 
instance, are the names of Tournefort to be excluded from 
use? In the breaking up of the Linnaean genus Lonicera, 
had not the Diervilla and Xylosteum (and if the division 
were to go farther, the Periclymenum and Caprifolium) of 
Tournefort, as well as the Symphoricarpos of Dillenius, an 
indisputable right to restoration ? Indeed Linnaeus was here 
plainly wrong in not adopting one of these prior names for 
the whole genus, instead of creating the new one. This, how- 
ever, was to be submitted to ; for, as Professor Agassiz re- 
marks, " the names sanctioned by Linnaeus are to be held as 
established above all others. Linnaeus, for instance, received 
very few genera of Echinodermata. Nowadays this class 
numbers many, among which some of those founded by Klein, 
Link and Breynius, long anterior to Linnaeus, hold their place 
with the modern ones of Lamarck, Miiller, etc. But no one 
now prefers that new names should be made for such genera, 
rather than that such approved anterior ones should be 
brought into use again. I certainly see no cause why we may 
not call to life the names of former authors when we divide 
the genera of Linnaeus." We think those naturalists blame- 



44 REVIEWS. 

worthy who do not. The third, fourth, and fifth of the 
British canons are accordant with Linnaean rules, and are 
regularly followed in botany. The next four relate to mat- 
ters which follow as a consequence of the law of priority ; 
but as to what relates to the use of synonymous names Pro- 
fessor Agassiz intimates that their rule is perhaps too ab- 
solute, and even contradictory to the Linnasan canon, § 244 : 
" Nomina generica, quamdiu synonyma digna in promptu 
sunt, nova non effingenda." 

The tenth rule, namely, " A name should be changed 
which has before been proposed for some other genus in 
zoology or botany, or for some other species in the same 
genus, when still retained for such genus or species," is not as 
well worded as the equivalent Linnsean canon, § 217, " Nomen 
genericum unum idemque ad diversa designanda genera as- 
sumptum, altero loco, excludendum erit." Mr. Agassiz re- 
marks, greatly to our surprise, that the enforcement of this 
rule would demand the sacrifice of almost half the generic 
names made in recent times. In our opinion, while the same 
names ought not to be given both in zoology and botany, 
the time is passed when received names are to be changed on 
this account. While writers in the different departments of 
zoology alone have doubly employed the same name " in ten 
thousand instances," we must see that cases of this sort be- 
tween zoologists and botanists, occupying such widely sepa- 
rated fields, are inevitable, at least until as perfect lists of 
zoological names shall be compiled and kept up as is done in 
botany. Besides, it is now utterly impossible for any single 
naturalist, or any joint committee of botanists and zoolo- 
gists, to determine, in half the cases that arise, whether a par- 
ticular genus is to be suppressed or retained in one depart- 
ment, so as to require or forbid a change of the posterior 
homonymous name in the other ; hence the practical applica- 
tion of the Linncean rule would now create tenfold more con- 
fusion than it can relieve. Each well-founded change of the 
sort does no more than obviate a possible inconvenience, 
while every needless one, in a genus of numerous species, 
draws after it a load of useless synonyms, which do not 



AGASSIZ' S ZOOLOGICAL NOMENCLATOR. 45 

serve, like genuine synonyms, to tell the history of the genus 
and mark the progress of our knowledge. The whole subject 
is forcibly presented by Professor Agassiz, in another section 
of his preface (p. xxviii et seq.}, where he states that he now 
knows three thousand generic names common to botany and 
zoology, which the Linnaean rule woidd require to be changed 
in one or the other department. But surely this number must 
comprise a host of synonyms long since laid on the shelf, as 
well as names of somewhat different formation or termination, 
although of the same derivation. In this case a small matter 
should give them impunity. If these changes must be made, 
no one could do the work for zoology better than Mr. Agassiz ; 
but he affirms it to be a task quite beyond his power, and 
justly concludes that, "in the present state of the science, 
generic names ought not to be changed solely on account 
of their being employed in both kingdoms of nature." To 
this conclusion the American Association evidently accede. 

As to generic names doubly or triply employed in the sev- 
eral classes of the animal kingdom (which, we are astonished 
to learn, already number nearly ten thousand), the neces- 
sity of applying the Linnaean canon is obvious, and would 
be imperative had not the evil reached such a height as to 
baffle remedy. The summum jus which demands the im- 
mediate change of nearly a moiety of the received zoological 
names would surely become summa injuria to the science, 
even if any naturalist were equal to the task of applying it. 
Justice must here be delayed in order that it may be rightly 
administered, and, as our author recommends, the business of 
gradually bringing this part of nomenclature under rule must 
be left to monographers and futui*e systematists. But let 
those upon whom the cacoethes nominandi is strong, obey 
our author's advice, desist from proposing new names in mere 
catalogues, and never attempt, while revising the genus which 
rightfully claims a particular name, to impose new names 
upon the homonymous genera in other classes, but leave that 
for their own respective monographers. It will be soon 
enough to give them new names, if such are needed, when 
the validity of these several genera is well made out. 



46 REVIEWS. 

Upon the 11th rule of the British Committee, namely, that 
" a name may be changed when it implies a false proposition 
which is likely to propagate important errors," Professor 
Agassiz remarks that the less this liberty is used the better, 
lest it should lead to licentiousness. 

The 12th rule ordains that " a name which has never been 
clearly defined in some published work should be changed for 
the earliest by which the object shall have been so defined." 
This law, our author remarks, " has become very necessary, 
since dealers in natural objects have begun to arrogate the 
authorship of books collected from catalogues, and demand 
that authors shall receive their names for dividing species. It 
is the same with names which remain unpublished in public 
or private collections, and to which the proprietors or cura- 
tors sometimes lay claim. But priority is to be conceded 
only to publication in a work which is accessible to the learned 
throughout the world. Yet while we strictly press the obser- 
vance of this law in respect to the publication by learned men 
of the results of their observations, so much the more must 
we brand with infamy those impudent parasites who prowl 
about museums to pick materials for their ojmscula, without 
mentioning the sources whence they have derived their spoil, 
and sometimes even furtively describing the species, the 
names of which they claim." The people alluded to well de- 
serve this censure. On the other hand, not less blameworthy 
are those who purposely pass by, instead of courteously adopt- 
ing, appropriate names under which naturalists often distri- 
bute their species in advance of publication. This felony is 
the more atrocious because remediless, and to be prevented 
by no rule except that of courtesy ; for the public good re- 
quires that priority should be conceded to actual publication 
alone. 

The two remaining laws (13th and 14th) are agreeable to, 
or identical with, Linnaean canons, and are approved by all 
good naturalists. 

The rules recommended by the British Committee for the 
future improvement of nomenclature are next considered; 
and as they are far the most commendable and in general use 



AGASSIZ'S ZOOLOGICAL NOMENCLATOR. 47 

among good naturalists, we shall only notice those that Mr. 
Agassiz criticises, or we have occasion to comment upon. The 
writer of the British Report has chosen to enforce the direc- 
tion, to avoid harsh and inelegant or sesquipedalian names, by 
citing, as an example of the kind, the " Enaliolimnosaurus 
crocodilocephaloides of a German naturalist ; " for which he 
is strongly censured by our author, who declares that no nat- 
uralist has ever proposed this name. Surely, if one is inclined 
" to cast stones into his neighbor's garden," as our author 
says, there is no lack of legitimate opportunity, nor necessity 
for fabricating hard names. 

The British Committee condemns the future employment 
of generic names which have been superseded by the rule of 
priority. But this is contrary to the canon, § 245 — " Nomen 
genericum unius generis, nisi supervacancum, in aliud trans- 
ferri non debet " (and to obs. under § 244), no less than to 
the practice of Linnaeus and of subsequent naturalists. For 
instance, Saururus of Plumier became a synonym of Piper, but 
this did not debar Linnaeus from the subsequent application 
of the name to a new genus. Sisyrinchium of Tournefort 
being included in Iris, Linnaeus gave the name to a different 
genus ; nor did he hesitate to adopt the genus which Ellis 
had dedicated to Hales, on account of an earlier Halesia of 
Browne, which had already sunk to a synonym. Why should 
a good name be forever tabooed in such cases, and why not, 
if occasion offers, allow it to be remarried to a new genus? 
We should be careful, however, not to reproduce names which 
are likely ever to be resuscitated in their former relation. 

The British Committee objects to the practice of giving to 
a genus the name which it bore as a species of a former 
genus. But, as Professor Agassiz justly remarks, when a 
species, which proves to be the type of a new genus, has a 
good proper name already, it seems quite as admissible to 
take that name for the genus and make a new one for the 
species, as to coin a new generic name, since either way a new 
name must be introduced : indeed it is preferable, because 
such Linnaean species frequently are found to comprise sev- 
eral, hitherto confounded, no one of which has a paramount 



48 REVIEWS. 

claim to the specific name : e. g. Cyprinus Gobio, C Leucis- 
cus, C Barbus, L. We go further, and maintain that the 
proper specific names are, ccetertbus paribus, always to be pre- 
ferred for genera in these cases, not only because that they 
are already familiar, but because they are most frequently 
old generic names which may claim under the law of priority. 
For example, Lonicera Diet-villa, l^. — Diervilla, Tourn. ; L. 
^'nnphoricarpos, L. = Symphoricarpos, Dill.; Rhamnus JPa- 
liurus, L,= Paliurus, Dod. ; R. Zizyphus, L. = Zizyphus, 
Dod. ; Rubus Dalibarda, L. = Dalibarda, L. ; and so of hun- 
dreds of proper specific names which have rightly resumed 
their generic rank. 

The next proposition of the British Committee, namely, that 
specific names, even when substantive or borrowed from per- 
sons or places, should uniformly be written with a small (in- 
stead of capital) initial, is so contrary to long usage and 
offensive to good taste, that we are surprised that it should 
anywhere find favor. Mr. Agassiz pointedly condemns it. 
The only reason assigned for the change is, that some people 
might not be able to distinguish the specific from the generic 
name without the aid of typography. But, as Dr. Gould has 
already remarked in this Journal, such persons would be 
misled by almost anything ; and the propounders of the rule 
should follow it consistently by writing their own cognomen 
with a small initial letter. We do not wonder that the Com- 
mittee of the American Association refused to reaffirm this 
rule, as applied to proper names from persons ; and we are 
quite sure that naturalists generally will not hesitate wholly 
to reject it; surely the committee would not approve the prac- 
tice of a late botanical author of this country, who reduced 
the proper scientific names of Linnaeus into adjective con- 
formity, by writing " Ranunculus jlammulus" instead of R. 
Flammula, "Thymus serpyllus" in the place of Thymus 
Serpyllum, and so on. 

Professor Agassiz severely condemns the proposition to re- 
strict the names of families to a uniform termination in idee, 
and their subdivisions to ince, without considering whether the 
words in question will receive that particular suffix kindly. 



AGASSIZ'S ZOOLOGICAL NOMENCLATOR. 49 

This is quite too straight-laced, and gives rise to many- 
awkward forms, or 

" Sesquipeclalia verba 
Vel nocitura sono, guttur lsesura loquentis," 

which is not worth while to encounter needlessly, for the sake 
of mere technical uniformity, at least when they may be 
avoided by some liberty of choice in the mode of prolonga- 
tion. 

The proposition, D. of the British Committee, which directs 
that the name of the original propounder of a species should 
adhere to it when transferred to a different genus, is warmly 
defended by some naturalists in England and, in a modified 
form, in our own country also. Few naturalists are now so 
well qualified to judge of the practical operation of this 
scheme as Professor Agassiz. He declares his opinion that, if 
received, " it will introduce horrible and remediless confu- 
sion," and that no possible multiplication of synonymy is 
likely to lead to so many difficulties as this new practice. He 
therefore strenuously opposes it by arguments drawn from the 
precepts and practice of Linnajus, who meant the specific 
name to be subordinate to the generic, and never intended it 
to be inferred that he who applied to a plant or animal a 
certain name was therefore its discoverer, or even its first 
systematic describer. He affirms that Linnaeus would have 
expressly rejected " Tyrannus crinitus, Linn, (s^;.)," were 
the innovation proposed in his day, and have written T. crini- 
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 than 
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 Jfuscicapa 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. crinitus, Linn. (s/>.) 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 MOHL'S VEGETABLE CELL. 51 

it is rather cumbrous, if it is to be used in every brief 
mention of the species, 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 

1 Grundziige der Anatomie und Physiologie der vegetabilischen Zelle. 
Hugo Von Mohl : Braunschweig, 1851. English translation (Principles 
of the Anatomy and Physiology of the Vegetable Cell) hy 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. 

" Air. 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 ] impose 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 MOHL'S 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 with 
imperfect 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 which 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 Halting 
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 
Collomia, the hairs of the achenium of Senecio, etc. (p. 18). 
The whole subject of spiral and other markings, rings, dots, 



5-i 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 Phaenogamous 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, 
mode 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 MOHL'S 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 the 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 when 
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 



56 REVIEWS. 

layers of wood are deposited, more fruit is perfected, this 
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 
loading 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, Mold 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 per- 
haps the principal part in its absorption and conveyance 
(p. 77). To the question, whether plants live on inorganic 
food alone or take in also organic matters, Mold 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 Mold, however, it is prov T ed 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., Mold 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. 86). "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 milk-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. 9G). 

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 Algm, 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 Algce (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 o\\v 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 Thrill ophytes ; but 
maintains that the small bodies, moving by two cilia, discov- 
ered by Decaisne and Thuret in the Fucacem^ 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. 1 Under the head of Reproduction 
by Seeds, Mold 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 wholly 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 Itzig'solm in Botanische Zeitung, May, 1850. — Here it is announced 
that the hlack dots 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, 1 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 spermatid. He unhesitatingly 
recognizes in them an apparatus of reproduction, doubtless analogous, at 
least in function, to those of the Florideous Algce, 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 Antherideset les Spores dequelques 
Fucus ; in Ann. Sci. Nat., 3 ser., iii. p. 5. — Here the corpuscles known 
to the earlier Algologists, and considered by Agardh and Montagne as a 
second kind of spores (a view which Mold adopts), are announced to be 
the spermatozoids of the antheridia of Fucaceai • 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 Alge'es 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. 
Nat., 3 ser., xiv. and xvi. (1850, 1851), with 30 plates. As to anther- 
idia, bodies like the free moving corpuscles of the Fucaceaz are shown 
likewise to occur in all the Florideoz, 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. 

LeVeiUe, 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. 

lislied 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 chapter, 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 
laro-e 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. Ward 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 Tricho-manes rad icons, which is ut- 
terly uncultivable 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. Recently, 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. A V here 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 growing 
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 with 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 1 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. Amoncr the conclusions or sujj- 
gestions that strike us as most true and timely are : The great 

1 Flora Indica : being a systematic account of the Plants of British 
India. Vol. I. By J. D. Hooker and Thomas Thomson. (Ranunculacea; 
to Fumariacea?.) 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 sys- 
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 mav 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 species, 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- 
pendent 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 generallv admitted as a 
fact, that a hybrid may readily be fertilized by the pollen of 
either of its parents ; and that if hybrid 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 they 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 occuiTence 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 offspring 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. 

A> 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. "We 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 
u P on 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 



BE 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 "Geographie Botanique" : of De Candolle 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 younger De Candolle 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 historj 7 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 Candolle 
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 

1 Geographie Botanique raisonnee, ou Exposition des Faites principauz et 
des Lois concernant la Bistribution Gt'ograpTiique des Plantes de V Europe 
Actuclle. Alphonse De Candolle. 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 aspect 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 temperature, 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 upon 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 
effected 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. 1 
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 vejreta- 
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 he 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 



BE CANBOLLE'S GfiOGRAPHIE BOTANIQUE. 71 

relative positions of land and sea may afford ; that many of 
these 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 operations 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 plants 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 we are inclined to venture a 
few remarks, as fitting opportunities occur. 



n 



HENFREYS BOTANY 

This is a well-planned, compact, and comprehensive w T ork, 
in which 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 ; with 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.) 



HENFREY'S 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 parts 
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 principles 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 that " if the previous 
education of medical students prepared them, as it should, 
with an elementary knowledge 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 materia 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 would 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, we should not take 
this trouble. 

As respects the first point noticed, our author, if wrong, 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 extremity, 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 monocoty- 
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 Cucurbitacece, or of a bean, 



HE NF RETS 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. 

Root-hairs or fibrillm 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- 
bitacece (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 " (p. 66). 
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 in 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 the 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 paleaceous 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 Rhamnacem 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 Gcra/ti'/c<<a', 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. c, to the superposition of the corres- 
ponding elements of successive whorls, — a view first sug- 
gested, we believe, by Lestibudois. 

" Real 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 Cructferce, we do not see 
what argument Megacarpwa 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 stiunens, but the contrary. 

The abnormal fertile flowers of Viola and Impatiens are 



HENFRETS BOTANY. 77 

not " achlamycleous," 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 Cactacece and some other cases. Professor 
Henfrey would seem to apply this view universally ; " for ex- 
ample, in Romcecv, Umbelliferce, Cucurbitacece, Compositce" 
[!] etc. But if applied to Rosa, why not to the Sanguisorbeoe, 
and to other Rosacece 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." 
We 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 
placentas 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 REVIEWS. 

rhaphe " (p. 130). No attentive student could fail to recog- 
nize the difference, especially in the families cited ( Cruciferce 
and Caryophyllacece). 

Ripening 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 pulp 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 with 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 oi'gans by a 
distinct substantive name, or in distinguishing by separate 
technical names fruits formed of a simple 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 diplotegia, while that of a Lily is called 
a capsule f And while we term the pod of Saxifraga stel- 
laris a capsule, and that of S. tridactylitcs 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 Labiatce are spoken of as 
seeds. 

As respects the systematic part, the chapters on the prin- 



HENFREY'S 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 were surprised at the statement that the bark and 
leaves of ffamamelis Virginica " 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 
JSamamelis 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 vields 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 the source of these extraordinary 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, amarine, 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. e., 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- 
diacece is an example of this prevalent misconception of what 
affinity is. This is more intelligible, however, than the " ap- 
proach " suggested of Aquifoliacece to Logayiiacece and Apo- 
cynacece, while their resemblance to Celastracece is thought 
to be of small account ; or that of Umbcllifcrce to IZubiacece, 
Saxifragacece, and even to Geraniacece, to which the resem- 
blances do indeed "seem rather superficial." Again, Xan- 
thoxylacece (i. e., Rutaccrp~) 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? 



H EN F KEY'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 writes 
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 iu 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 we 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 " 1 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 dcs Plantes. These forms our author reduces to six 
species, and the alimentaiy sorts in cultivation to three, namely, 
Cucurbita maxima, C. 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, 0. perennis and C. d'ujitata, 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. Melopepo ; and C. JPepo, 
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 birthplace of these plants. He remarks that 
C. maxima and C. moschata have been known in European 
gardens scarcely above two centuries ; but that C. Pepo was 
perhaps known to the Greeks and the Romans in the time of 
Pliny. 

1 American Journal of Science and Arts, 2 ser., xxiv. 440. 



84 RE VIE WS. 

The younger De Candolle, in his discussion of the history 
and origin of the principal cultivated plants, which forms 
a most interesting chapter of his " Ge'ographie 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 
drawing 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, 1G13, 4to ; also " Voyages et Decouvertes 
faites en la Xouvelle France depuis l'annee 1615, jusques a la 
fin de l'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, Citrouittes 
and Courges are repeatedly mentioned, along with maize 
{Bled d'lnde) and beans ; c. g. : 

" Nous y vismes force dtrouittes^ courges & petum, qu'ils 
cultiuet aussi. . . . Pour les febues elles comec^oiet a cntrer 
en fleur, come faysoyet les courges et citrouilles " (p. 68). 

" Ceux que nous auions 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 conime coucombres, qui sont tres- 
bonnes " (p. 77). 

See also pp. 83, 115, 116. Of course it does not follow 
that tliese 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 
Cucurbita verrucosa. This common squash is, according to 
Naudin, a variety of C. JPepo, as also is C. aurantia (the 
C. Texana vel ovifera, Gray, "PI. Liudheimerianse"), 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 Romans 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 Cucurbita Pcpo, as that 
species is now understood. According to De Candolle, 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, lie 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 pcrennis and C. digitata, to which we add 
that C. Pepo itself (i. e., C. ovifera or aurantia) grows wild 
in the same district with Cperennh, 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 Cucurbitacece, 
and especially that the species of Cucurbita, cross-breed with 
extreme facility. According to Naudin this is true of the races 
only inter sc. 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 Plantes (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 

coining 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 URTICACEiE. 

Dr. Weddeli/s preliminary studies upon the proper Urti- 
cttcece 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 1 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), 1856-57. (American Journal of Science and 
Arts, 2 ser., xxv. 109.) 



S s REVIEWS. 

the " Archives " which contain that model one on the Mal- 
pighiacece of his lamented botanical master. It illustrates in 
detail about 470 species, under 40 genera, and is accompanied 
by twenty 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 Polypetaloe, our 
author searches among 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 Malvacece are the highest development. According to 
Weddell's happy illustration, Malvaceae crown the summit of 
a three-sided pyramid, with Sterculiacece, Byftncriacece, and 
Tiliacece just below them, one upon each face ; under the 
Byttncri a cc<v he ranks the Euphorbiacece with the Antides- 
ineee, 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 Euphorbiacece, he inscribes the 
Urticacece, with the Cupuliferce 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 Ilaloragcce, Caryophyllacece 
including Uhcebrece), as others do through a series of two or 
three orders, or even as the same group does {c. g., Caryo- 
phyllacece) through a series of orders on the other side of the 
pyramid. 



WEDDELL'S MONOGRAPH OF URTICACEJE. 89 

The reason why this mode of representation 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 arrangement even of its great 
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 we 
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 TJrtl- 
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 Jl/tlcacece 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 Plica pumila. Dr. Weddell has 
overlooked the fact that Rafinesque 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. AVeddell 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 Boehnieria but with the peni- 
cillate stigma of Urtica, namely, his Hesperoenide tenella 
("Pacific Railroad Reports," iv. p. 139). This little plant, it 
now appears, comes nearest to Wight's monotypic genus Cha- 
mabaina 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 



RADLKOFERS PROCESS OF FECUNDATION IN THE 
VEGETABLE KINGDOM. 1 

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, 
Cohn, 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 
Vaucher 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 Befruchlungsprocess im PJlanzenreiche. L. Radlkofer. Leipsic, 
1857. (English translation by Arthur Henfrey in Annales 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 Bulbochsete, and especially in Sphceroplea, 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. Coining to rest, these 
zoospores germinate, by elongation and the formation of trans- 
verse partitions, into adult tln-ead-like plants, consisting of a 



FECUNDATION IN THE VEGETABLE KINGDOM. 93 

row of cells. In Sphseroplea 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 ; Vou- 
cher's old observations to this effect having been confirmed 
by Braun and Pringsheim. 

That in the Facacece or olive-green Algm 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 perfectly 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 Linnasan Society nearly three 
years ago, this fine memoir 1 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 Balanophorece, by J. D. Hooker, (sep- 
arately issued from the Transactions of the Linnaian Society of London, 
xxii.) London, 1857. (American Journal of Science and Arts, 2 ser., 
xxv. 116.) 



DR. HOOKER ON THE BALANOPHOREJE. 95 

elaborate plates which 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 organograpkically 
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 BalanopJiorece 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 Balanof)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 phamogamous. 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, Martius, Blume, and others, 
who describe them as a sporuliferous 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 Helosis and Langsdorffia, 'mycelio Fungorum 
quam maxime analogum.' 4. The resemblance between the 
articulated filaments on the capitula of the Helosideoe and the 
paraphyses of Musci ; and between the pistils of Balano- 
p7wrece 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 Rafflesiacece) 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 Balanojjhorece, then, as degraded members of the 
Dicotyledonous class, Dr. Hooker follows Brown and Griffith 
in regarding Raffle siaeccti as near to Aristolochi acece, and in 
denying all affinity between these and Bolanojihorece. 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 BALANOPHOREJE. 97 

as those may be termed which 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 applied in the present case leads Dr. Hooker to 
the conclusion that the nearest relatives of Balanophorece are 
the Heloragece, a group itself, " consisting for the most part 
of reduced forms of Onagrariece" 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 Onagracea, 
while it is impossible to sever the cbain of evidence winch 
binds the genus to Loudonia and Ilaloragis. 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 Orbcmchece, 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 Santalacece into Loran- 
thacece. 

It is quite probable that our author would deny the degra- 
dation in the latter case, judging from some points which he 
makes when considering whether the group of Balanophorece, 
" putting aside any consideration of its relationship with other 



98 REVIEWS. 

orders, and regarding it per sr, . . . 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 jjer 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 style 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 " Balanophorece 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 
it 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 special j>rovision 
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 comjmct, their extent as 



DR. HOOKER ON THE BALANOPHOREJE. 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 
type 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 deeply than our author 
the want of fixed and philosophical principles for the subor- 
dination of characters and the study of affinities 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 Balanophorece, if our author has demon- 
strated anything, it is that they belong to the highest class of 
plants, 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, 1 
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; a?id (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 plants 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, produced in three 
months of vegetation, contained 8.444 of carbon derived from 
the carbonic acid of the air and 0.1G66 grams of nitrogen. 
The 1.4 grams of nitrate of potash supplied 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 water 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. argophyllus 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 " 

Xo. 4 " " " 0.16 " 

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 grains 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, 31G.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. 108 

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 specimens 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 manui-e, 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 Vesle 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 how active, beyond all ordinary conception, must the 
process of nitrification be all over the land ; and how vast the 
supply of assimilable nitrogen for the use of vegetation ! 



BEXTHAM'S HAND-BOOK OF THE BRITISH FLORA. 

Oxe of the best systematic botanists — of the soundest 
judgment and the largest expei*ience, both in European and 
exotic botany — has deemed it no unfit employment of a por- 
tion of his valuable time to prepare a volume 1 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 De Can- 
dolle's " Flora Francaise," 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- 
cpaired for the purpose in view, and " the omission, in nu- 
merous instances, of microscopical, anatomical, or theoretical 

1 Hand-book of (lie British Flora ; a Description of the Flowering Plants 
and Ferns indigenous to, or naturalized in, the British Isles : for the use of 
Beginners and Amateurs. By George Bentnam. 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 descrip- 
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), Mr. 
Bentham has no superior. 4. The geographical rauge 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 Linnseus, 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 



10G REVIEWS. 

be understood without a more complete acquaintance with 
trifling, vague, 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 well as in this country, and on 
repeated comparison of specimens collected from the most 
varied and distant points of the geographical areas of the 
several species." 

G. 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; between 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 five, 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 which 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 Linnsean 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. g., 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 disposed 
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 Limosella. 

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 Myriophyllum ; 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 ? 

Poly carp (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 not 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 Linncean 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 l 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 or altera- 
tion, as in other cases, c'est la premiere r 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 Considerations 
sur VHeredite dans les Vegetaux. Par M. Louis Vilmorin. Paris, 1859. 
(American Journal of Science and Arts, 2 ser., xxvii. 440.) 



HO REVIEWS. 

plant exhibits little flexibility. To detect the 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 Vilmorin 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 will 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 offsjH'ing to parents. Xow 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, Loth 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 REVIEWS. 

It is in this 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 parent, 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, 1 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. Xuttall first named 
and described it nearly thirty years ago ; and, while he re- 
ferred it to Sesleria, suspected it to be sui generis, and threw 

1 Two new Genera of Dioecious Grasses of the United Slates. 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 OhlorideoB ; 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 axilliflora 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 
specific 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,. Ktli. — 
Xutt. Sesleria, Nutt. Gen. i. p. 65." But neither Kunth nor 
any other author has described a genus Calanthera. We 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 manuscript 
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 Hartwegiana?," Mr. Bentham applies the 
name of " Lasiostega humilis, Rupprecht (ined.) " to No. 
250, which he had before called a Triodia. 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 spike- 
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 paleae 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 pistilliferous, 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 1 
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, 
1860. (American Journal of Science and Arts, 2 ser., xxx. 275.) 



110 HE VIEWS. 

Clematis Virginiana over the leaf. (We venture to add, in 
passing, C. I iorna 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 Men- 
ziesia globularis 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 Mhodo- 
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 diffei'ent, having light and pendent branches, 
when well grown forming broad and thick masses, and loaded 
with its handsome rose-colored blossoms. While Leucothoe 
Catesbcei is called " a very pretty shrub," the far hand- 
somer Andromeda floribunda, so much prized by our nursery- 
men, gets no commendation. Magnolia 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 anywbei*e in North Carolina. The flowers of 
J/, a j rdata are described as if larger than those of M. Fra- 
seri^ instead of the contrary ; we could hardly say much for 
their beauty, except in comparison to those of the common 
Cucumber-tree. Primus 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 (Abies alba~) 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 lias pretty 



BENTBAM'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 Report will pass to a second edition, — when we 
hope it will be largely augmented. 



BENTHAM'S FLORA OF HONGKONG. 

The present work 1 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 Hongkongensis ; 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 plant 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 Floi-as. 

" 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 with thickened leaves, such as 
those of a Houseleek, a Meseinbryanthemum, 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, Quincuncial 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 perigynous 
and epigynous (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 



BENTHAWS FLORA OF HONGKONG. 121 

adhesion reaches 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 166, 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 hvpogseous 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 atteuded, 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 5 , 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 principal 
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. Paragraph 
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 phsenogaraous 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 



12-4 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 s]3ecies, 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. AY as 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 southern 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 beuds 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 
temperatures 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, where 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 may, in- 
deed, reach to the pole itself. Phamogamic 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 TTalden 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 phsenogamous 
species peculiar to the arctic zone, and only one peculiar 
genus, namely, the grass Pleuropogon. 1 Of the TG2 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 hemisphere, 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 species, 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 adopted, as to know 
clearly what the adopted scale is. Here we are not left in 

1 Douglasia is mentioned in another place (p. 2G9) 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 Brande'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 762 
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 appropriate to his particular jrarpose, 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 which 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 by keeping 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 comparatively recent geo- 
logical epoch, and that immediately previous to the glacial 
period or since then" (p. 279). All we could 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 Rhinanihus minor may well be considered as "not a 
sufficiently constant form to rank as a race even," while Limo- 
sella ten ui folia could rank for no more than a race, and while 
Castilleia septentrionalis and C. pallida, 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 PediculaHs lanata, Willd., does not 
rightfully merge in P. hirsuta this side of the glacial period, 
although it perhaps may into P. Pangsdorffii, 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 upon 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. 



ALPIIONSE DE CANDOLLE ON THE VARIATION 
AND DISTRIBUTION OF SPECIES. 

This is the title of a paper 1 by Monsieur De Candolle grow- 
ing out of his study of the Oaks. It was published in the 
November number of the " Bibliotheque 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 

1 Etude sur I'Espece, a Voccasion d'une Revision de la Famille des Cupu- 
lifers. Par M. Alphonse De Candolle. (American Journal of Science 
and Arts, 2 ser., xxxv. 431.) 



VARIATION AND DISTRIBUTION OF SPECIES. 131 

for their bulk, offered small hope of novelty. The subject 
was both extremely trite and extremely difficult. Happily 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, inter 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 
Epiontology} all lead up to and converge into this class of 
questions, while recent theories shape and point the discus- 
sion. So we look with eager interest to see what light 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 Candolle proposes for 
the study of the succession of organized beings, to comprehend, therefore, 
palaeontology and everything included under what is called geographical 
botany and geographical zoology, — the whole forming a science parallel 
to geology, — the latter devoted to the history of unorganized bodies, 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 palaeontology ; since ontology, the science of being, has an estab- 
lished meaning as referring to mental existence, — i. e., 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 development, some- 
times irrespective of such relations oi 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, the 
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 each pavtic- 



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- 
elas Ilex, Q. cocci/era, Q. acutifolia, 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- Tlcqmxahuensis to Q. Sartor it of Liehmann, since these 
two differ only in their entire or their toothed leaves. From the 
fact that the length of the peduncle varies in Q. Robur and many 
other Oaks, I might have combined Q. Seema nnii, Liebm., with 
Q. salicifolia, Ne'e. I have not admitted these inductions, but have 
demanded visible proof in each particular case. Many species are 
thus left as provisional ; but in proceeding 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, do 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 Candolle 
pursue in the case — of every-day occurrences to most work- 
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 he 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 Cupvliferce 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." 






VARIATION 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 asDe 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 species, 
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 systematise " 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. Robur, with twenty-eight varieties, all spon- 
taneous. Of Q. Lusitanica eleven varieties are enumerated, 
of Q. Calliprinos 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 Cupuliferce, 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 
effects 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 Coniferce and Birches), 
the Common Oak, to some extent, and the Turkey Oak de- 
cidedly, are diminishing and retreating, and this wholly 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 reasonings. 

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 disposition 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. Quercus Robar 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 Linnaeus this Common Oak of Europe was all of one 
species. But of late years the greater number of European 
botanists have regarded it as including three species, Q.pedun- 
culata, Q. sessiliflora, and Q. pubescens. De Candolle looks 
with satisfaction to the independent conclusion which he 
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 Plantarmn." 
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. Itobur, which 
De Candolle recognizes, all but six, he remarks, fall naturally 
under the three sub-species, jyechinadata, sessiliflora, and 
jiubescens, 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 worked out by Dr. 
Hooker two or three years ago. 1 

A special advantage of the Cupuliferce 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 all 
experience. From such considerations, i. 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 Cupvliferce of the Old World attained their actual ex- 
tension before the pi*esent separation of Sicily, Sardinia and 
Corsica, or of P>ritain, from the European continent. 
1 Natural History Review, Jan., 18G2. 



VARIATION AND DISTRIBUTION OF SPECIES. 139 

This view once adopted, and this course once entered upon, 
has to be pursued farther. Quercus Robur 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. Robur, 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. Robur on the one hand, and from those of 
Q. Douglasii, 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. Robur, traceable in Europe up to 
the commencement of the present epoch, looks eastward and 
far into the past on far distant shores. 

Q. Ilex, the Evergreen Oak of southern Europe and north- 
ern Africa, reveals a similar archaeology ; but its presence in 
Algeria leads De Candolle to regard it as a much more 
ancient denizen of Europe than Q. Robur ; and a Tertiary 
Oak, Q. ilicoides, from a very old Miocene bed in Switzer- 
land, is thought 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 Mexican rela- 
tives, may probably enough be regarded as early offshoots 
from the same stock with Q. Rex. 

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 REVIEWS. 

former times they were as liable to variation as they now are ; 
ami 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 De 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 upon 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 the 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 " Geographie Botanique," 
the multiple hypothesis, although in principle not abandoned, 
is seen to lose its point, in view of the probable high antiquity 
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 extraordinary, 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, in winch 
the permanence of species was ably maintained, with his treat- 
ment of the same subject in a work just issued in England, 
which, however, has not yet reached us. 

A belief in the derivation of species may be maintained 
along with a conviction of great persistence of specific char- 
acters. This is the idea of the excellent Swiss vegetable palae- 
ontologist Heer, who imagines a sudden change of specific 
type at certain periods, and perhaps is that of Pictet. Fal- 
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 which 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 c 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. 143 

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. With him, 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 he correct, they 
seem clearly to indicate that the older elephants of Europe, such as 
E. meridionalis and E. antiquus, were not the stocks from which 
the later species, E. primigenius and E. Africanus, 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 with 
the Miocene E. planifrons of India ; and of E. primigenius, with 
the existing India species. 

" Another reflexion is equally strong in my mind, — that the 
means which have heen adduced to explain 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 around 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, 
nowever different from another, can he 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 Mammalia, 
fossil and recent, cannot furnish a species which has had a wider 
geographical distribution, and passed through a longer term of time, 
and through more extreme changes of climatal conditions, than the 
Mammoth. If species are so unstable, and so susceptible of muta- 
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 cotemporaries, 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 Duorao 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 ensemble 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 phyllotaxis, 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. 145 

least idea as to how the change from the one to the other 
comes to pass. But it is interesting, and in this connection 
perhaps instructive, to remark that, while some dicotyledonous 
plants hold to the verticillate, l. e., opposite-leaved phyllotaxis 
throughout, a larger number — through the operation of some 
deep-seated and innate principle, which we cannot fathom — 
change abruptly into the other species at the second or third 
node, and change back again in the flower, or else effect a 
synthesis of the two species in a manner which is puzzling to 
understand. Here is a change from one fixed law to another, 
as unaccountable, if not as great, as from one specific form 
to another. 

An elaborate paper on the vegetation of the Tertiary period 
in the southeast of France, by Count Gaston de Saporta, piib- 
lished in " Annales des Sciences Naturelles," xvi. pp. 309- 
344, — which we have not space to analyze, — is worthy of 
attention from the general inquirer, on account of its analysis 
of the Tertiary flora into its separate types, Cretaceous, Aus- 
tral, Tropical, and Boreal, each of which has its separate and 
different history, — and for the announcement that " the hia- 
tus, which, in the idea of most geologists, intervened between 
the close of the Cretaceous and the beginning of the Tertiary, 
appears to have had no existence, so far as concerns the vege- 
tation ; that in general it was not by means of a total over- 
throw, followed by a complete new emission of species, that 
the flora has been renewed at each successive period ; and 
that while the plants of southern Europe inherited from the 
Cretaceous period more or less rapidly disappeared, as also 
the austral forms, and later the tropical types (except the 
Laurel, the Myrtle, and the Chamcerops humilis), the boreal 
types, coming later, survived all the others, and now compose, 
either in Europe, or in the north of Asia, or in North America, 
the basis of the actual arborescent vegetation." Especially 
" a very considerable number of forms nearly identical with 
Tertiary forms now exist in America, where they have found, 
more easily than in our [European] soil, — less vast and less 
extended southward, — refuge from ulterior revolutions." The 
extinction of species is attributed to two kinds of causes : the 



146 REVIEWS. 

one material or physical, whether slow or rapid ; the other 
inherent in the nature of organic beings, incessant, but slow, 
in a manner latent, but somehow assigning to the species, as 
to the individuals, a limited period of existence, and, in some 
equally mysterious but wholly natural way, connected with 
the development of organic types : — "By type meaning a col- 
lection of vegetable forms constructed upon the same plan of 
organization, of which they reproduce the essential lineaments 
with certain secondary modifications, and which appear to run 
back to a common point of departure." 

In this community of types, no less than in the community 
of certain existing species, Saporta recognizes a prolonged 
material union between North America and Europe in former 
times. Most naturalists and geologists reason in the same 
wa y» — som e more cautiously than others, — yet perhaps most 
of them seem not to perceive how far such inferences imply 
the doctrine of the common origin of related species. 

For obvious reasons such doctrines are likely to find more 
favor with botanists than with zoologists. But with both the 
advance in this direction is seen to have been rapid and great ; 
yet to us not unexpected. We note, also, an evident disposi- 
tion, notwithstanding some endeavors to the contrary, to allow 
derivative hypotheses to stand or fall upon their own merits, 
— to have indeed upon philosophical grounds certain pre- 
sumjitions in their favor, — and to be, perhaps, quite as cap- 
able of being turned to good account as to bad account in 
natural theology. 1 

1 What the Rev. Principal Tulloch remarks in respect to the philosophy 
of miracles has a pertinent application here. We quote at second hand : 

" The stoutest advocates of interference can mean nothing more than 
that the Supreme Will has so moved the hidden springs of nature that a 
new issue arises on given circumstances. The ordinary issue is supplanted 
by a higher issue. The essential facts before us are a certain set of phe- 
nomena, and a Higher Will moving them. How moving them ? is a ques- 
tion for human definition ; the answer to which does not and cannot affect 
the divine meaning of the change. Yet when we reflect that this Higher 
A\ ill is everywhere reason and wisdom, it seems a justcr as well as a more 
comprehensive view to regard it as operating by subordination and evolu- 
tion, rather than by interference or violation." 



VARIATION AND DISTRIBUTION OF SPECIES. 147 

Among the leading naturalists, indeed, such views — taken 
in the widest sense — have one and, so far as we are now 
aware, only one thorough-going and thoroughly consistent 
opponent, namely, Mr. Agassiz. 

Most naturalists take into their very conception of a species, 
explicitly or by implication, the notion of a material connec- 
tion resulting from the descent of the individuals composing 
it from a common stock, of local origin. Mr. Agassiz wholly 
eliminates community of descent from his idea of species, and 
even conceives a species to have been as numerous in individ- 
uals, and as widespread over space, or as segregated in dis- 
continuous spaces, from the first as at a later period. 

The station which it inhabits, therefore, is with other natu- 
ralists in nowise essential to the species, and may not have 
been the region of its origin. In Mr. Agassiz's view the hab- 
itat is supposed to mark the origin, and to be a part of the 
character of the species. The habitat is not merely the place 
where it is, but a part of what it is. 

Most naturalists recognize varieties of species ; and many, 
like De Candolle, have come to conclude that varieties of the 
highest grade, or races, so far partake of the characteristics of 
species, and are so far governed by the same laws, that it is 
often very difficult to draw a clear and certain distinction be- 
tween the two. Mr. Agassiz will not allow that varieties or 
races exist in nature, apart from man's agency. 

Most naturalists believe that the origin of species is super- 
natural, their dispersion or particular geographical area, nat- 
ural, and their extinction, when they disappear, also the result 
of physical causes. In the view of Mr. Agassiz, if rightly 
understood, all three are equally independent of physical cause 
and effect, are equally supernatural. 

In comparing preceding periods with the present and with 
each other, most naturalists and palaeontologists now appear 
to recognize a certain number of species as having survived 
from one epoch to the next, or even through more than one 
formation, especially from the Tertiary into the Post-tertiary 
period, and from that to the present age. Mr. Agassiz is 
understood to believe in total extinctions and total new crea- 



148 REVIEWS. 

tions at each successive epoch, and even to recognize no exist- 
ing species as ever contemporary with extinct ones, except in 
the case of recent exterminations. 

These peculiar views, if sustained, will effectually dispose of 
every form of derivative hypothesis. 

Returning for a moment to De Candolle's article, we are dis- 
posed to notice his criticism of Linnaeus's " definition " of the 
term species (" Philosophia Botanica," No. 157) : " Species 
tot numeramus quot diversae forniae in principio sunt creatae," 
— which he declares illogical, inapplicable, and the worst that 
has been propounded. " So, to determine if a form is specific, 
it is necessary to go back to its origin, which is impossible. 
A definition by a character which can never be verified is no 
definition at all." 

2s ow, as Linnaeus practically applied the idea of species with 
a sagacity which has never been surpassed, and rarely equalled, 
and indeed may be said to have fixed its received meaning in 
natural history, it may well be inferred that in the phrase 
above cited he did not so much undertake to frame a logical 
definition, as to set forth the idea which, in his opinion, lay 
at the foundation of species. On which basis A. L. Jussieu 
did construct a logical definition : " nunc rectius definitur pe- 
rennis individuorum similium successio continuata generatione 
renascentium." The fundamental idea of species, we would 
still maintain, is that of a chain, of which genetically-con- 
nected individuals are the links. That, in the practical rec- 
ognition of species, the essential characteristic has to be in- 
ferred, is no great objection, — the general fact that like 
engenders like being an induction from a vast number of in- 
stances, and the only assumption being that of the uniformity 
of nature. The idea of gravitation, that of the atomic consti- 
tution of matter, and the like, equally have to be verified in- 
ferentially. If we still hold to the idea of Linnaeus, and of 
Agassiz, that existing species were created independently, and 
essentially all at once at the beginning of the present era, we 
could not better the propositions of Linnaeus and of Jussieu. 
If, on the other hand, the time has come in which we may accept, 
with De Candolle, their successive origination, at the com- 



VARIATION AND DISTRIBUTION OF SPECIES. 149 

mencement of the present era or before, and even by deriva- 
tion from other forms, then the " in principio " of Linnaeus 
will refer to that time, whenever it was, and his proposition be 
as sound and wise as ever. 

In his "Geographie Botanique " (ii. pp. 1068-1077) De 
Candolle discusses this subject at length, and in the same 
interest. Remarking that of the two great facts of species, 
nainely, likeness among the individuals, and genealogical con- 
nection, zoologists have generally preferred the latter, 1 while 
botanists have been divided in opinion, he pronounces for the 
former as the essential thing, in the following: argumentative 
statement : — 

" Quant a moi, j'ai e'te' conduit, dans ma definition de l'espece, k 
mettre decidement la resseiublance au-dessus des caracteres de suc- 
cession. Ce n'est pas seulement a cause des circonstances propres au 
regne ve'ge'tal, dont je m'occupe exclusivement ; ce n'est pas non plus 
afin de sortir ma definition des theories et de la rendre le plus pos- 
sible utile aux naturalistes descripteurs et nomenclateurs, c'est aussi 
par un motif philosophique. En toute chose il faut aller au fond des 
questions, quand on le peut. Or, pourquoi la reproduction est-elle 
possible, habituelle, fe*conde indefiniment, entre des etres organises 
que nous dirons de la meme espece ? Parce qu'ils se ressemblent et 
uniquement a cause de cela. Lorsque deux especes ne peuvent, ou, 
s'il s'agit d'aniinaux superieurs, ne peuvent et ne veulent se croiser, 
c'est qu'elles sont tres differentes. Si Ton obtient des croisements, 
c'est que les individus sont analogues ; si ces croisements donnent des 
produits feconds. c'est que les individus etaient plus analogues ; si 
ces produits eux-memes sont fe'conds, c'est que la ressemblance etait 
plus grande ; s'ils sont fecond habituellement et indefiniment, c'est 
que la ressemblance inteVieure et ext^rieure e"tait tres grande. Ainsi 
le degre" de ressemblance est le fond ; la reproduction en est seule- 
ment la manifestation et la mesure, et il est logique de placer la 
cause au-dessus de l'effet." 

"We are not at all convinced. "We still hold that genealog- 
ical connection, rather than mutual resemblance, is the funda- 
mental thing, — first on the ground of fact, and then from the 

1 Particularly citing Flourens : " La ressemblance n'est qu'une condi- 
tion secondaire ; la condition essentielle est la descendance : ce n'est pas 
la ressemblance, c'est la succession des individus, qui fait l'espece." 



150 REVIEWS. 

philosophy of the case. Practically, no botanist can say what 
amount of dissimilarity is compatible with unity of species ; 
in wild plants it is sometimes very great, in cultivated races, 
often enormous. De Candolle himself informs us that the dif- 
ferent variations which the same Oak tree exhibits are signifi- 
cant indications of a disj)osition to set up separate varieties, 
which, becoming hereditary, may constitute a race ; he evi- 
dently looks upon the extreme forms, say of Quercus Robar, 
as having thus originated ; and on this ground, inferred from 
transitional forms, and not from their mutual resemblance, as 
we suppose, he includes them in that species. This will be 
more apparent should the discovery of the transitions, which 
he leads us to expect, hereafter cause the four provisional 
species which attend Q. Robur to be merged in that species. 
It may rightly be replied that this conclusion would be arrived 
at from the likeness step by step in the series of forms ; but 
the cause of the likeness here is obvious. And this brings in 
our " motif philosophique." 

Not to insist that the likeness is after all the variable, not 
the constant, element, — to learn which is the essential thing, 
resemblance among the individuals or their genetic connec- 
tion, we have only to ask which can be the cause of the other. 

In hermaphrodite plants (the normal case), and even as 
the question is ingeniously put by De Candolle in the above 
extract, the former surely cannot be the cause of the latter, 
though it may, in case of crossing, offer occasion. But, on 
the ground of the most fundamental of all things in the con- 
stitution of plants and animals, " the fact incapable of farther 
analysis, that individuals reproduce their like, that character- 
istics are inheritable," the likeness is a direct natural conse- 
quence of the genetic succession, — and it is logical to place 
the cause above the effect. 

We are equally disposed to combat a proposition of De 
Candolle's about genera, elaborately argued in the " Geographie 
Botanique," and incidentally re-affirmed in his present article, 
namely, that genera are more natural than species, and are 
more correctly distinguished by people in general, as is shown 
by vernacular names. But we have no space left in which to 
present some evidence to the contrary. 



DR. HOOKER ON WELWITSCHIA. 151 

Here we must abruptly close our long exposition of a paper 
which, from the scientific position, ability, and impartiality of 
its author, is likely at this time to produce a marked impres- 
sion. We would also direct attention to an earlier article in 
the same important periodical (namely, in the " Bibliotheque 
Universelle " for May, 1862), on the European Flora and the 
Configuration of Continents in the Tertiary Epoch, a most 
interesting abstract of, and commentary on, the introductory 
part of Heer's " Flora Tertiaria Helvetiae," as reedited and 
translated into the French by Gaudin, with additions by the 
author. 



DR. HOOKER ON WELWITSCHIA. 

This is a separate issue, in folio form, of a memoir 1 in the 
current (24) volume of the Transactions of the Linnsean 
Society of London, illustrated by fourteen superb and elabo- 
rate plates, the expense of which has been mainly defrayed by 
the Royal Society, from a parliamentary fund placed at its 
disposal for the promotion of scientific research. By the co- 
operation of these two learned societies, the fruits of Dr. 
Hooker's admirable researches are given to the scientific 
world in a form and manner worthy of them and of the won- 
derful subject. 

A good idea of the vegetable wonder in question is given in 
the following brief account of its appearance and prominent 
characters, drawn partly from the descriptions of its discoverer, 
and partly from specimens sent to England : — 

" The Welwitschia is a woody plant, said to attain a century 
in duration, with an obconic trunk about two feet long, of 
which a few inches rise above the soil, presenting the appear- 
ance of a flat, two-lobed depressed mass, sometimes (according 
to Dr. Welwitsch) attaining fourteen feet in circumference (!), 
and looking like a round table. When full grown it is dark 
brown, hard, and cracked over the whole surface (much like 

1 On the Welwitschia, a new genus of Gnetacece. By Joseph Dalton 
Hooker. (American Journal of Science and Arts, 2 ser., xxxvi. 434.) 



152 REVIEWS. 

the burnt crust of a loaf of bread) ; the lower portion forms 
a stout tap-root, buried in the soil, and branching- downward 
at the end. From deep grooves in the circumference of the 
depressed mass two enormous leaves are given off, each six 
feet long when full grown, one corresponding to each lobe ; 
these are quite fiat, linear, very leathery, and are split to the 
base into innumerable thongs that lie curling upon the sur- 
face of the soil. Its discoverer describes these same two 
leaves as being present from the earliest condition of the 
plant, and assures me that they are in fact developed from the 
two cotyledons of the seed, and are persistent, being replaced 
by no others. From the circumference of the tabular mass, 
above, but close to the insertion of the leaves, spring stout, 
dichotomously branched cymes, nearly a foot high, bearing 
small erect scarlet cones, which eventually become oblong, and 
attain the size of those of the common Spruce Fir. The scales 
of the cones are very closely imbricated, and contain, when 
young and still very small, solitary flowers, which in some 
cones are hermaphrodite (structurally but not functionally), 
in others female. The hermaphrodite flower consists of a 
perianth of four pieces, six monadelphous stamens, with glo- 
bose trilocular anthers, surrounding a central ovule, the inte- 
gument of which is produced into a styliform sigmoid tube, 
terminated by a discoid apex. The female flower consists of 
a solitary erect ovule, contained in a compressed utricular 
perianth. The mature cone is tetragonous, and contains a 
broadly winged fruit in each scale. Its discoverer observes 
that the whole plant exudes a resin, and that it is called 
Tumbo by the natives, — whence he suggests that it may bear 
the generic name of Tumboa; but this he withdrew at my 
suggestion, for reasons which I shall presently give. It in- 
habits the elevated sandy plateau near Cape Negro (lat. 15° 
40' S.) on the southwest coast of Africa." 

Welwitschia mirabilis, Hook, fil., was also detected and 
made known — indeed the first actual materials, with a draw- 
ing of the plant, were sent to England — by Mr. Baines from 
the Damara country, in lat. 24° or 24° S., about 500 miles 
south of Cape Negro. Mr. Baines is an artist, and his original 



DR. HOOKER ON WELWITSCHIA. 153 

colored sketch of a plant in fruit is reproduced on the first 
plate of the memoir. It appears as if five-leaved ; but prob- 
ably one of the two original leaves is split in two, and the 
other into three segments. As might be inferred from the 
form and structure, the AVelwitschia inhabits a dry region. 
Mr. Monteiro writes to Dr. Hooker : — 

"... About thirty miles distant from the coast, I passed 
a plain about three miles across, on which this plant was 
growing abundantly ; that is to say, I saw about thirty speci- 
mens on my line of march. The plain was perfectly dry, and 
bare of other vegetation than the AVelwitschia and a little 
short grass. The ground was a hard quartzose schist. The 
Welwitschia was generally growing near the little ruts worn 
in the plain by running water during tbe rainy season." 

And from Damara Land, Mr. Anderson writes that, — 

" Kain rarely or never falls where this plant exists. (Yet 
the night dews are heavy, as other authorities mention.) I 
have crossed and recrossed Damara Land throughout its en- 
tire length and breadth, but only found the plant growing on 
that desperately arid flat, stretching far and wide about Wal- 
visch Bay." 

We are familiar with plants of very diverse orders of Di- 
cotyledons and Monocotyledons which are adapted to arid 
regions by great restriction of surface. Here a similar plan 
is adopted by a Gymnosperm ; for the resemblance to Coni- 
fers and CasuarinecB indicated by Dr. "Welwitsch is shown 
by Dr. Hooker to import a close affinity, the author referring 
the plant to Gnetacece near to Ephedra. Its permanently 
abbreviated ascending axis — of which the greater part con- 
sists of the first internode, below the cotyledons — increases 
in thickness but hardly in length, develops no other than the 
seminal pair of leaves, above which the disciform bilobate 
axile portion or " crown," gradually produced, bears year by 
year only leafless inflorescence. 

Haemanthus equally bears a pair of leaves ; but these die 
off as the season of drought advances, when the plant is 
reduced to a minimum of surface in its spherical bulb, — 
which outspreads a new pair of leaves when the rainy season 



154 BE VIEWS. 

returns. But in Welwitscliia the two leaves are permanent. 
Wherefore they are firm and coriaceous, and, increasing by 
basal growth from year to year, the older parts doubtless 
become inactive at length, while fresh surface below is an- 
nually renewed, under the shelter (as Dr. Hooker describes) 
afforded by the deep grooves which in old plants separate the 
growth of the hypocotyledonary stock from that of the crown 
above, and is filled by the tender growing bases of the leaves. 
Having given a detailed generic character of Welwitscliia 
and a comparative view of the Gnetaceous genera, now three 
in number, Dr. Hooker proceeds to describe at length the 
trunk, leaves, inflorescence, flowers, fertilization, embryogeny, 
and seeds of this curious subject, — comparing it, in the latter 
respects, with Coniferce and Cycadacece on the one hand, and 
with Santalum and Loranthus on the other, and closing with 
a general summary of the results. 

An abstract or analysis of this most important paper is 
beyond our present reach and space. But we may refer to 
some of the special points. 

The most obvious peculiarity of Welwitscliia is, that " it 
appears to be the only perennial flowering plant which at no 
period has other vegetative organs than those proper to the 
embryo itself, — the main axis being represented by the rad- 
icle, which becomes a gigantic caulicle, and develops a root 
from its base and inflorescences from its plum alary end, and 
the leaves being the two cotyledons in a very highly developed 
and specialized condition." It is an excellent case, accord- 
ingly, if any such were still needed for showing the nature 
of the radicle as stem, or ascending axis (not root), — a view 
which we supposed observation had long ago demonstrated. 
Dr. Hooker, in a note, refers to this view as expressed by 
Adr. de Jussieu in his " Cours Elcmentaire " (which appeared 
in 1843 and 1844), and in Gray's Introduction to Botany 
(" Botanical Text Book "), 1858. But the same view is taken 
in all the earlier editions of the latter work ; even in the first 
(1842) the radicle is spoken of as the first internode of the 
stem (p. 28, note) ; and probably the idea will be found dis- 
tinctly expressed in works of an earlier date. Dr. Hooker, 



DR. HOOKER ON WELW1TSCHIA. 155 

in the note referred to, assents to the proposition that " the 
radicle is rightly regarded as an axis," i. e., an ascending 
axis, " and not a root," but does not agree that it is an in- 
ternode. To us, the one implies the other. Conceiving, as 
we do, the fundamental idea of the morphology of the phse- 
noganious plant to be, that the ascending axis consists of a 
series of superposed internodes, each crowned by a leaf-bear- 
ing point or ring (the node), the first internode must needs 
be that which is crowned by the first leaf or pair of leaves, 
the cotyledons ; and its whole development confirms this view. 

Dr. Hooker notes the curious fact that in Welwitschia 
flower-buds are occasionally produced on the stock below the 
insertion of the leaves, that is on the radicle or caulicle it- 
self ; and Dr. Masters pointed out to him analogous cases of 
shoots thus originating, one of which was described by Bern- 
hardi thirty years ago. It is simply the case of adventitious 
buds ; these might seem as likely to occur on the first inter- 
node as on any later one. 

Welwitschia, having a dicotyledonous embryo, has also 
essentially an exogenous stem, i. e., "the vascular system is 
referable to the exogenous plan, but its arrangement into 
concentric wood wedges is very rude." But the superadded 
isolated and closed vascular bundles of the stock and root, and 
especially their losing themselves in the periphery of the 
stock, are endogenous analogies. So also is the strictly par- 
allel and free venation of the leaves ; yet, as there are no 
cross veinlets, thus favoring the splitting up of the leaf into 
lacinise, this looks as much or more towards Cycadacem and 
broad-leaved Coniferce. 

The total absence of anastomosing veinlets in the leaf, each 
nerve representing a single and independent vascular axis, 
extending, in Welwitschia, from the axis of the trunk to the 
apex of the leaf, causes such leaves as these and those of 
Dammara, etc., to " resemble more closely a series of parallel 
uninerved leaves united by cellular tissue, than a foliar ex- 
pansion of parenchyma traversed by one system of inosculat- 
ing vessels, and the frequent presence of many linear coty- 
ledons in these plants seems to favor this view, as does the 



156 REVIEWS. 

mixed character of the foliage of Podocarpus, of which some 
species have uninerved and others many-nerved leaves. The 
numerous flower-buds along the periphery of the crown also 
further favor this view." That is, in Welwitschia, where 
this ingenious surmise carries a plausibility, which it does not 
when applied to Podocarpus. 

The binary symmetry of Welwitschia, beginning with the 
cotyledons, is carried through the inflorescence up to the de- 
cussating pairs of bracts of the cones and the two leaflets in 
each whorl of the hermaphrodite perianth. But the stamens 
are six, at first sight a monocotyledonous analogy ; yet they 
may be regarded three sets of two, notwithstanding their 
monadelphy. The flowers are dioecio-polygamous, i. e., of 
two sorts, one female, the other structurally hermaphrodite, 
but the gyna3cium sterile, though well-developed, except that 
no embryo-sac appears. The hermaphrodite cones and their 
flowers accord in many respects strikingly with the male cones 
of Ephedra ; but the anthers are trilocular, which is remark- 
able. The simple ellipsoidal pollen is the same in both. In 
Ephedra the stamens vary from two to eight, and the column 
is solid, there being no rudiment of a gynaecium. 

The female fruitful cones are about three inches long, and 
bright red when fresh. 

The integument of the ovule, as in Gnetum, is prolonged 
at the summit into a style-like body, thus closely simulating 
a pistil ; and the apex of this styliform tube, which is thin 
and merely erose in the fertile flower, in the structurally 
hermaphrodite flower is dilated into a broad papillose disk, 
exactly imitating a highly developed stigma — a marked in- 
stance of a highly developed organ which is functionless ; for 
no pollen has been detected upon it, and no embryo-sac in 
the nucleus. Here Dr. Hooker speculates upon " the pos- 
sibility of Welwitschia being the only known representative 
of an existing or extinct race of plants, in which such a 
stigma-like organ was really capable of performing the func- 
tion of a stigma. And, when we see this organ occurring in 
a hermaphrodite flower, it is easy to suppose that we have in 
Welwitschia a transition in function, as well as in structure, 



DR. HOOKER ON WELW1TSCHIA. 157 

between the gynmospermous and angiospermous Dicotyledons ; 
and that the ideal race consisted of hermaphrodite plants, in 
which the office of the stigma of the carpellary leaf was per- 
formed by a stigmatic dilatation of the ovular coat itself." 

This assumes that the gymnospermous theory established 
by Brown is correct (whatever be the nature of the cone- 
scales, rameal or carpellary, simple or compound), and ap- 
plicable to the Gnetaceous as well as to the Coniferous type. 
This view, lately much questioned, Dr. Hooker maintains, 
and enforces, as respects Gnetacece, by very convincing and 
in part wholly original arguments, drawn from his own re- 
searches upon the present plant and its allies. We refer to 
pp. 28 to 31, which we could not readily condense, and have 
not room to copy. The same is to be said in regard to the 
resemblances or analogies in gynsecial structure between 
Gnetacece and Loranthacece, etc., — a subject upon which we 
await expectantly Professor Oliver's investigations. More- 
over, as Dr. Hooker remarks upon another page, the decisive 
or final comparative view of the structures in question cannot 
be had until the homology of the ovule itself is settled. In 
cases where the flower is so simplified that the nucleus of 
an ovule directly terminates the floriferous axis, and is sur- 
rounded by few and simple, or peculiarly specialized, invest- 
ments, the discrimination of these must be difficult enough, 
and must ultimately depend upon the theory adopted as to 
the nature of the ovular coats. If these be regarded as foliar 
(as a rigid application of adopted morphological principles 
will require), then a complete transition from gymnospermy 
to angiospermy is probable, and may be expected to be de- 
monstrable. 

The fertilization and embryology of Welwitschia have been 
wonderfully worked out, considering the materials, by Dr. 
Hooker, and the two most elaborate and valuable plates of 
the memoir are filled with the details. Suffice it to say, that 
it appears that the pollen must be brought by insects to the 
ovule of the female flowers, at an early period, before the 
nucleus is covered by the ovular coat or by the perianth, and 
before the former has produced its styliform apex, down 



158 REVIEWS. 

which it would be nearly impossible to convey the grains of 
pollen which were bodily found on the nucleus, with their 
tubes there produced. So that, notwithstanding the carpel-like 
form of the ovule, the impregnation is absolutely gymno- 
spermous. As to embryo-formation also, " there is a general 
agreement in many most essential particulars with Cyca- 
daceoB and Coniferce" especially, and beyond what has 
already been adverted to, in "the free enibryo-sac being 
filled with endosperm-cells previous to fertilization, the nu- 
merous secondary embryo-sacs, the position of the germinal 
vesicle at the base of these sacs, and in the high development 
of the long tortuous susj)ensor." There is an agreement with 
Angiosperms, however, in several particulars, especially in 
that of " the germinal vesicle giving rise to one embryo only." 
And it is concluded that, in special reference to Sentalum 
and Loranthus, " Welwitschia presents an embryogenic pro- 
cess intermediate between that of Gymnosperms and An- 
giosperms." 

And here we should not omit to mention that its wood 
differs from that of all known Gymnosperms in wanting the 
disc-bearing wood-cells ! 

It will be conceded that "Welwitschia is " the most won- 
derful discovery, in a botanical point of view, that has been 
brought to light during the present century." Also, that Dr. 
Hooker has enjoyed (and improved) an opportunity une- 
qualled by any botanist since that which placed the Rafflesia 
in Mr. Brown's hands. 



DARWIN'S MOVEMENTS AND HABITS OF CLIMBING 

PLANTS. 

This is a long paper 1 read before the Linnaean Society 
in February last. The investigations which it records were 
made, we believe, during a period when the author's ordinary 

1 On the Movements and Habits of Climbing Plants. By Charles Darwin. 
Journal Linnsean Society of London, ix. London, 18G7. (American Jour- 
nal of Science and Arts, 2 ser., xl. 273 ; xli. 125.) 



HABITS OF CLIMBING PLANTS. 159 

scientific labors were interrupted by illness, — as was no less 
the case with respect to his former papers on Dimorphous and 
Trimorphous Flowers and his volume on the Fertilization of 
Orchids by the aid of Insects. Of these works and of the 
present, — side issues as they are, — it may fairly be said, that 
they show a genius for biological investigation, and a power 
of turning common materials and ordinary observations to high 
scientific account, which, if equalled, have not been surpassed 
since the days of Hunter and Charles Bell. This will be the 
opinion equally, we suppose, of those who favor and of those 
who dislike Mr. Darwin's theory of the gradual transforma- 
tion of specific forms through natural selection, upon which, 
indeed, all these collateral researches have a bearing, direct or 
incidental. In the present case the bearing is obvious. The 
gradual acquisition by certain plants of advantageous pecu- 
liarities is inferred from the gradation of forms and functions. 
Properties and powers which are latent or feebly developed 
in most plants are taken advantage of by some, made specially 
useful, and enhanced from generation to generation. Tendril- 
bearing plants, the most specialized in structure and the most 
exquisitely adapted to the end in view, are supposed to have 
been derived from leaf-climbers, and these in turn from simple 
twiners. 

The author states that he was led to this subject by a brief 
note, communicated to the American Academy in the summer 
of 1858 (and reprinted in this Journal), in which the writer 
of the present notice recorded his observation of the coiling 
of certain tendrils by a visible movement promptly following 
an extraneous irritation. Mr. Darwin's observations were 
more than half completed before he became aware that the 
spontaneous revolution of the stems and of some tendrils of 
climbing plants had been observed and recorded almost forty 
years ago, and nearly at the same time, by Palm and by Von 
Mohl, and had been the subject of two memoirs by Dutrochet, 
published more than twenty years ago. But the mode in 
which the free and growing end of a stem sweeps around 
seems not to have been previously well made out, having been 
more or less confounded with the torsion of the axis which 



100 REVIEWS. 

many twining stems, such as those of the Hop and the Morn- 
ing Glory, are apt to undergo. It is plain to see, however, 
that many stems which revolve do not twist at all ; and those 
that do never could twist on their axis at every revolution 
without speedy destruction, — indeed usually do not twist until 
they have ceased revolving. Eveiy one must have noticed that 
the growing extremity of a Hop, Convolvulus, or other twiner, 
when unsupported, hangs over or stretches out horizontally to 
one side. But it is not so well known that this outstretched 
portion, while at the proper age, is continually sweeping 
round, in circles widening as it grows, and always in the same 
direction, in search of some object round which to twine. The 
Hop revolves with the sun ; the Convolvulus, Bean (Phaseo- 
lus), etc., against the sun, that is, in the same directions that 
they twine. Two or three inte modes are usually revolving at 
the same time. Mr. Darwin observed thirty-seven revolutions 
in one internode of a Hop, — the first revolution made in about 
twenty-four hours, the second in nine hours, the third and the 
following ones up to the eighth in a little over three hours 
each. " The shoot had now grown 3| inches in length, and 
carried at its extremity a young internode an inch in length, 
which showed slight changes in its curvature. The next or 
ninth revolution was effected in two hours and thirty minutes. 
From this time forward the revolutions were easily observed. 
The thirty-sixth revolution was performed at the usual rate ; 
so was the last or thirty-seventh, but it was not quite com- 
pleted ; for the internode abruptly became upright, and after 
moving; to the centre became motionless. I tied a weight to 
its upper end, so as slightly to bow it, and thus to detect any 
movement ; but there was noue. Some time before the last 
revolution the lower part of the internode had ceased to move. 
... It moved during five days ; but the more rapid move- 
ment after the third revolution lasted during three days and 
twenty hours. The regular revolutions from the 9th to the 
36th inclusive, were performed at the average rate of 2 h. 
31 m. The w r eather was cold, and this affected the tempera- 
ture of the room, especially during the night, and consequently 
retarded a little the rate of movement. . . . After the seven- 



HABITS OF CLIMBING PLANTS. 161 

teenth revolution the internocle liad grown from 1| to 6 inches 
in length, and carried an internocle 1| inch long, which was 
just perceptibly moving ; and this carried a very minute ulti- 
mate internode. After the 21st revolution the penultimate 
internode was 2|- inches long, and probably revolved in a 
period of about three hours. At the 27th revolution our 
lower internode was 8|, the penultimate 3|-, and the ultimate 
2| inches in length ; and the inclination of the whole shoot 
was such that a circle 19 inches in diameter was swept by it. 
When the movement ceased the lower internode was 9 and 
the penultimate G inches in length ; so that, from the 27th to 
the 37th revolutions inclusive, three internodes were at the 
same time revolving." — (pp. 3, 4.) 

The shoots of many climbers sweep their circles more 
rapidly than the Hop, the common Pole Bean (Phaseolus 
vulgaris) in rather less than two hours, Convolvuluses of va- 
rious species in the same time or rather less ; while more 
woody stems naturally move more slowly, some requiring from 
24 to 50 hours for each revolution. But the thickness or tex- 
ture of the shoot does not govern the rate, many slender shoots 
moving slower than some stout ones, and some lignescent 
quicker than other purely herbaceous ones. The movement 
appears to be accelerated, up to a certain point, by raising the 
temperature, or rather is retarded by lowering it ; but while 
the conditions are nearly the same, the rate is often remark- 
ably uniform. The quickest rate of revolution of a proper 
stem observed by Mr. Darwin was that of a Scyph ran thus, in 
77 minutes. When the light comes from one side, the semi- 
circle towards the light is usually described in less time, often 
in less than half the time, of that from the light. The ten- 
dency of young stems to turn toward the light is here active 
as usual, but is overcome by a superior force. The end of 
the shoot describes circles or broad ellipses, or else, from in- 
sufficient power or mechanical disadvantages, narrow ellipses, 
semicircles, or irregular figures. A horizontal shoot of con- 
siderable length will thus be found, not unfrequently, to sway 
from side to side in a semicircular course, while the extreme 
internodes are making complete revolutions. 



162 REVIEWS. 

A striking illustration of the amount of space that may be 
swept over is afforded by a case in which Mr. Darwin allowed 
the top of a Ceropegia to grow out almost horizontally to the 
length of 31 inches, — three long internodes terminated by two 
short ones. The whole revolved at rates between 5^ and 6|- 
hours for each revolution, the tip sweeping a circle of above 
five feet in diameter and 16 feet in circumference, traveling 
therefore at the rate of at least 32 inches per hour. " It was 
an interesting spectacle to watch the long shoot sweeping, 
night and day, this grand circle, in search of some object 
around which to twine." 

As to the nature of this revolving movement, Mr. Darwin 
clearly shows that it is not a torsion of the axis, but a succes- 
sive bending (similar to that by which ordinary stems bend 
toward the light), the direction of which is constantly and 
uniformly changing. " If a colored streak be painted (this 
was done with a large number of twining plants) along, we 
will say, the convex line of surface, this colored streak will, 
after a time depending on the rate of revolution, be found to 
lie along one side of the bow, then along the concave side, 
then on the opposite side, and, lastly, again on the opposite 
convex surface. This clearly proves that the internodes, dur- 
ing the revolving movement, become bowed in every direction. 
The movement is, in fact, a continuous self-bowing of the 
whole shoot, successively directed to all points of the com- 
pass." It is an automatic movement, of the same character 
as those which these and other parts of plants effect in chang- 
ing position or direction, sometimes slowly and sometimes with 
a visible motion. The movement may be likened in one case 
to that of the hour-hand or the minute-hand of a clock, in the 
other to the second-hand, but in both is as truly a vital move- 
ment as is the contraction of an involuntary muscle. It must 
be effected — as Mr. Darwin recognizes — either by the con- 
traction of the cells on the concave side, or by the turgescence 
and elongation of those on the convex side of the internode, 
or by both, — probably the former, as various facts go to show ; 
but questions of that kind are not investigated in the present 
essay. 



HABITS OF CLIMBING PLANTS. 163 

No differences in this regard are observable in the behavior 
of exogenous or endogenous stems, or even of those of climb- 
ing Ferns. Lygodium scandcns, according to Darwin, re- 
volves like other twiners ; it completes its revolutions in six 
hours, or on a very hot day in five (moving against the sun, 
which is much the commoner case) ; this is about the average 
rate of Phaenogamous twiners, like which it comports itself 
in all respects. Our own L. 2)almatum, we find, revolves in 
the same way, in about four hours, the temperature being 75° 
Fahr. 

The power of revolving depends, of course, upon the gen- 
eral health and vigor of the plant, and upon the age of the 
shoot, is retarded by lowering the temperature, is interrupted 
by any considerable disturbance, such as exposure to cold or 
to much jarring ; carrying the plant from one place to another, 
or cutting off a shoot and placing it in water, stops the move- 
ment for a time, just as it does the more vivid automatic 
movement of Desmodium gyrans. But each internode is so 
independent that cutting off an upper one does not affect the 
revolutions of the one beneath. Twining stems are far from 
being insensible to the action of light (as Mohl supposed), 
the half-revolution toward the light being not uncommonly 
twice faster than that from it ; but as the rate of revolution 
by day and by night is nearly the same, the one half of the 
circuit is accelerated just as much as the other is retarded. 
This influence of the light is quite remarkable when we con- 
sider the slenderness of most revolving internodes, the small 
surfaces they expose, and that their leaves are little developed. 

The design, as we must term it, of this revolving of the end 
of twining stems is obvious, and usually effectual. Such 
stems, even when no supporting object is within their reach, 
will reach each other, and by twining together make a mutual 
support, from which, as they lengthen, they may reach yet 
farther. The connection of the revolving with twining is 
obvious, though the latter is not a necessary consequence ; for 
many stems revolve which do not twine, but climb in some 
other way. 

" When at last the [revolving] shoot meets with a support, 



1<34 REVIEWS. 

the motion at the point of contact is necessarily arrested, but 
the free projecting part goes on revolving. Almost immedi- 
ately another and upper point of the shoot is brought into 
contact with the support and is arrested ; and so onward to the 
extremity of the shoot ; and thus it winds round its support. 
When the shoot follows the sun in its revolving course, it 
winds itself round the support from right to left, the support 
being supposed to stand in front of the beholder ; when the 
shoot revolves in an opposite direction, the line of winding is 
reversed. As each internode loses from age its power of 
revolving, it loses its power of spirally twining round a sup- 
port. If a man swings a rope round his head, and the end 
hits a stick, it will coil round the stick according to the direc- 
tion of the swinging rope ; so it is with twining plants, the 
continued contraction or turgescence of the cells alono- the 
free part of the shoot replacing the momentum of each atom 
of the free end of the rope. 

" All the authors, except Von Mohl, who have discussed the 
spiral twining of plants maintain that such plants have a nat- 
ural tendency to grow spirally. Mohl believes (S. 112) that 
twining stems have a dull kind of irritability, so that they 
bend toward any object which they touch. Even before 
reading MohFs interesting treatise, this view seemed to me 
so probable that I tested it in every way that I could, but 
always with negative results. I rubbed many shoots much 
harder than is necessary to excite movement in any tendril or 
in any foot-stalk of a leaf-climber, but without result. I then 
tied a very light forked twig to a shoot of a Hop, a Ceropegia, 
Sphaerostema, and Adhatoda, so that the fork pressed on one 
side alone of the shoot and revolved with it ; I purposely 
selected some very slow revolvers, as it seemed most likely 
that these would profit from possessing irritability ; but in no 
case was any effect produced. Moreover, when a shoot winds 
round a support, the movement is always slower, as we shall 
immediately see, than whilst it revolves freely and touches 
nothing. Hence I conclude that twining stems are not irri- 
table ; and indeed it is not probable that it should be so, as 
nature always economizes her means, and irritability would be 



HABITS OF CLIMBING PLANTS. 165 

superfluous. Nevertheless I do not wish to assert that they 
are never irritable ; for the growing axis of the leaf-climbing, 
but not spirally twining, Lophospermum scandens is, as we 
shall hereafter see, certainly irritable ; but this case gives rue 
confidence that ordinary twiners do not possess this quality, 
for directly after putting a stick to the Lophospermum, I saw 
that it behaved differently from any true twiner or any other 
leaf -climber. 

" The belief that twiners have a natural tendency to grow 
spirally probably arose from their assuming this form when 
wound round a support, and from the extremity, even whilst 
remaining free, sometimes assuming this same form. The free 
internodes of vigorously growing plants, when they cease to 
revolve, become straight, and show no tendency to be spiral ; 
but when any shoot has nearly ceased to grow, or when the 
plant is unhealthy, the extremity does occasionally become 
spiral. I have seen this in a remarkable degree with the ends 
of the shoots of the Stauntonia and of the allied Akebia, 
which became closely wound up spirally, just like a tendril, 
especially after the small, ill-formed leaves had perished. The 
explanation of this fact is, I believe, that the lower parts of 
such terminal internodes very gradually and successively lose 
their power of movement, whilst the portions just above move 
onward, and in their turn become motionless ; and this ends in 
forming an irregular spire. 

" When a revolving shoot strikes a stick, it winds round it 
rather more slowly than it revolves. For instance, a shoot of 
the Ceropegia took 9 hours and 30 minutes to make one com- 
plete spire round a stick, whilst it revolved in 6 hours ; Aris- 
tolochia gig as revolved in about 5 hours, but took 9 hours and 
15 minutes to complete its spire. This, I presume, is due to 
the continued disturbance of the moving force by its arrest- 
ment at each successive point ; we shall hereafter see that even 
shaking a plant retards the revolving movement. The terminal 
internodes of a long, much-inclined, revolving shoot of the 
Ceropegia, after they had wound round a stick, always slipped 
up it, so as to render the spire more open than it was at first ; 
and this was evidently due to the force which caused the revolu- 



1G6 REVIEWS. 

tions being now almost freed from the constraint of gravity, 
and allowed to act freely. "With the Wistaria, on the other 
hand, a long, horizontal shoot wound itself at first, in a very 
close spire, which remained unchanged ; but subsequently, as 
the shoot grew, it made a much more open spire. With all 
the many plants which were allowed freely to ascend a sup- 
port, the terminal internodes made at first a close spire ; and 
this, during windy weather, well served to keep the shoots in 
contact with their support ; but as the penultimate internodes 
grew in length, they pushed themselves up for a considerable 
space (ascertained by colored marks on the shoot and on the 
support) round the stick, and the spire became more open. 

" If a stick which has arrested a revolving shoot, but has 
not as jet been wound round, be suddenly taken away, the 
shoot generally springs forward, showing that it has continued 
to press against the stick. If the stick, shortly after having 
been wound round, be withdrawn, the shoot retains for a time 
its spiral form, then straightens itself, and again commences 
to revolve. The long, much-inclined shoot of the Ceropegia 
previously alluded to offered some curious peculiarities. The 
lower and older internodes, which continued to revolve, had 
become so stiff that they were incapable, on repeated trials, of 
twining round a thin stick, showing that the power of move- 
ment was retained after flexibility had been lost. I then 
moved the stick to a greater distance, so that it was struck by 
a point 2 j inches from the extremity of the penultimate inter- 
node ; and it was then neatly wound round by this part and 
by the ultimate internode. After leaving the spirally wound 
shoot for eleven hours, I quietly withdrew the stick, and in 
the course of the day the curled part straightened itself and 
re-commenced revolving ; but the lower and not curled portion 
of the penultimate internode did not move, a sort of hinge 
separating the moving and the motionless part of the same 
internode. After a few days, however, I found that the 
lower part of this internode had likewise recovered its revolv- 
ing power. These several facts show, that, in the arrested 
portion of a revolving shoot, the power of movement is not 
immediately lost, and that when temporarily lost it can be re- 



HABITS OF CLIMBING PLANTS. 167 

covered. When a shoot has remained for a considerable time 
wound round its support, it permanently retains its spiral 
form even when the support is removed. 

" When a stick was placed so as to arrest the lower and rigid 
internodes of the Ceropegia at the distance at first of 15 and 
then of 21 inches from the centre of revolution, the shoot slowly 
and gradually slid up the stick, so as to become more and more 
highly inclined ; and then, after an interval sufficient to have 
allowed of a semi-revolution, it suddenly bounded from the 
stick and fell over to the opposite side, to its ordinary slight 
inclination. It now recommenced revolving in its usual 
course, so that after a semi-revolution it again came into con- 
tact with the stick, again slid up it, and again bounded from 
it. This movement of the shoot had a very odd appearance, 
as if it were disgusted with its failure but resolved to try 
again. We shall, I think, understand this movement by con- 
sidering the former illustration of the sapling, in which the 
contracting surface was supposed to creep from the southern, 
by the eastern, to the northern, and thence back again by the 
western side to the southern face, successively bowing the sap- 
ling in all directions. Now with the Ceropegia, the stick 
being placed a very little to the east of due south of the plant, 
the eastern contraction could produce no effect beyond pressing 
the rigid internode against the stick ; but as soon as the con- 
traction of the northern face began, it would slowly drag the 
shoot up the stick ; and then, as soon as the western contrac- 
tion had well begun, the shoot would be drawn from the stick, 
and its weight coinciding with the northwestern contraction, 
would cause it suddenly to fall to the opposite side with its 
proper slightly inclined positions ; and the ordinary revolving 
movement would go on. I have described this case because it 
first made me understand the order in which, the contracting 
or turgescent cells of revolving shoots must act. 

" The view just given further explains, as I believe, a fact 
observed by Von Mohl (S. 135), namely, that a revolving 
shoot, though it will twine round an object as thin as a thread, 
cannot do so round a thick support. I placed some long 
revolving shoots of a Wistaria close to a post between 5 and 



168 REVIEWS. 

G inches in diameter, but they could not, though aided by me 
in many ways, wind round it. This apparently is owing to 
the flexure of the shoot, when winding round an object so 
gently curved as this post, not being sufficient to hold the 
shoot to its place when the contracting force creeps round to 
the opposite surface of the shoot ; so that it is at each revolu- 
tion withdrawn from its support." — (pp. 9-13, passim.') 

The successive shifting of the contracting side of the shoot, 
which explains the revolution or bowing in turn in every direc- 
tion, no less explains the twining round a proper support, 
leaving however some idiosyncrasies unexplained. Some ten- 
drils and some petioles of leaf-climbing plants equally possess 
this revolving power ; but their usefulness depends mainly 
upon additional and more special endowments, — mainly upon 
the power of directly responding by curvature to the contact, 
more or less prolonged, of an extraneous body. 

Of Leaf-climbers, no instance is more familiar than that of 
Clematis or Virgin's Bower. Little more was known of them 
than that they climbed by curling their petioles (common or 
partial) around neighboring objects. Mr. Darwin made obser- 
vations upon eight species of Clematis, seven of Tropseolum, 
the common species of Maurandia, Lophospermum, Fumaria, 
etc., as also upon Gloriosa and Flagellaria, which climb by a 
tendril-like production of the tip of the leaf. From the sum- 
mary it appears that plants which belong to eight families are 
known to have clasping petioles, and those of four families 
climb by the tips of their leaves. In almost all of them the 
young internodes revolve, in some of them as extensively as 
in twining plants, — the movement being plainly serviceable 
in bringing the petioles or the tips of the leaves into contact 
with surrounding objects. Those whose shoots revolve most 
freely are also capable of twining spirally around a support ; 
but when the stem twines (as in Clematis Sieboldii and C. 
calycina, but not in C. Viticella), it has the peculiarity of wind- 
ing first in one direction for two or three turns, and then in the 
opposite direction. The petioles are principally efficient in these 
plants, and that by means of an endowment which is not shown 
to belong to twining stems, with one or two exceptions. That 



HABITS OF CLIMBING PLANTS. 169 

is, the petioles or their divisions are sensitive to the contact of 
an extraneous body, contracting on the side touched so as to 
curve or coil around it. That the footstalk is directly sensi- 
tive to the touch, just as tendrils are, Mr. Darwin proved by 
lightly rubbing them with a twig for a few times, when in the 
course of some hours it bends to the rubbed side, afterwards 
becoming straight again ; or by leaving the body in contact it 
is permanently clasped by the footstalk. So sensitive are some 
footstalks that " a loop of thread weighing a quarter of a grain 
caused them to bend ; a loop weighing one-eighth of a grain 
sometimes acted, and sometimes not." In one instance, in 
Clematis Flammula, even the sixteenth part of a grain caused 
a petiole to bend through nearly 90 degrees. With rare 
exceptions only the young petioles are sensitive. Take the 
cultivated Clematis Viticella for an illustration of the mode in 
which the leaves do the work of climbing. 

"The leaves are of large size. There are three pairs of 
lateral leaflets and a terminal one, all borne by rather long 
petioles. The main petiole bends a little, angularly, down- 
ward at each point where a pair of leaflets arises, and the 
petiole of the terminal leaflet is bent downward at right angles ; 
hence the whole petiole, with its rectangularly bent extremity, 
acts as a hook. This, with the lateral petioles directed a little 
upward, forms an excellent grappling apparatus by which the 
leaves readily become entangled with surrounding objects. If 
they catch nothing, the whole petiole ultimately grows straight. 
Both the medial and lateral petioles are sensitive ; and the 
three branches, into which the basi-lateral petioles are gener- 
ally subdivided, likewise are sensitive. The basal portion of 
the main petiole, between the stem and the first pair of leaf- 
lets, is less sensitive than the remainder, but it will clasp a 
stick when in contact. On the other hand, the inferior sur- 
face of the rectangularly bent terminal portion (carrying the 
terminal leaflet), which forms the inner side of the end of the 
hook, is the most sensitive part ; and this portion is manifestly 
best adapted to catch distant supports. To show the differ- 
ence in sensibility, I gently placed loops of string of the same 
weight (in one instance weighing .82 of a grain) on the sev- 



170 REVIEWS. 

eral lateral and on the terminal sub-petioles ; in a few hours 
the latter were bent, but after 24 hours no effect was produced 
on any of the lateral petioles. Again, a terminal sub-petiole 
placed in contact with a thin stick became sensibly curved in 
45 minutes, and in 1 hour 10 minutes had moved through 
ninety degrees, whereas a lateral petiole did not become sen- 
sibly curved until 3 hours and 30 minutes had elapsed. In 
this latter case, and in all other such cases, if the sticks be 
taken away, the petioles continue to move during many hours 
afterward ; so they do after a slight rubbing ; but ultimately, 
if the flexure has not been very great or long-continued, they 
become, after about a day's interval, straight again." — (p. 31.) 
In numerous cases, notably in Solanum jasminoides, the 
petiole when clasped increases very greatly in thickness and 
rigidity, undergoing a change in its woody structure by which 
the fibro-vascular bundles, originally semi-lunar in cross-sec- 
tion, develop into a closed ring, like that of an exogenous 
stem. 

Lophospermum scandens of the gardens climbs, like its 
allies Maurandia and Rhodochiton, by clasping petioles ; but 
in this plant, alone, the young internodes are also sensitive to 
the touch. 

" When a petiole clasps a stick, it draws the base of the 
internode against it ; and then the internode itself bends 
toward the stick, which is thus caught between the stem and 
the petiole as by a pair of pincers. The internode straight- 
ens itself again, excepting the part in contact with the stick. 
Young internodes alone are sensitive, and these are sensitive 
on all sides along their whole length. I made fifteen trials 
by lightly rubbing two or three times with a thin twig several 
internodes ; and in about 2 hours, but in one case in 3 hours, 
all became bent ; they became straight again in about 4 hours, 
subsequently. An internode, which was rubbed as much as 
six or seven times with a twig, became just perceptibly curved 
in 1 hour 15 minutes, and subsequently in 3 hours the curva- 
ture increased much ; the internode became straight again in 
the course of the night. I rubbed some internodes one day on 
one side, and the next day on the opposite side or at right 



HABITS OF CLIMBING PLANTS. 171 

angles ; and the curvature was always toward the rubbed 
side." 

Here, then, is one case in which the sensibility of a stem is 
manifest, and is turned to useful account. The peduncles of 
the allied Maurandia semperflorens are also sensitive and 
flexuous, although Mr. Darwin insists that they are useless 
for climbing. That some stems should be sensitive might 
have been expected ; for tendrils of axial nature (e. qr. of 
Passiflora gracilis) are not less sensitive than those of foliar 
nature, as of Legmn inosce, Cucurhitacece, and Cobaea. And 
if twining: stems in general are not endowed with " a dull 
kind of irritability," as Mohl conjectured, it may well be 
because the equally wonderful automatic revolving movement 
leaves no need for it; In general, the most striking cases of 
automatic movement belong to leaves or their homologues. 

The distinction can be only somewhat arbitrarily drawn 
between Leaf-climbers — especially those with small or un- 
developed leaflets, or where the tip of the leaf forms a hook 
or tendril-like projection — and Tendril-climbers. The ten- 
dril, however, whether answering to leaf or stem, is the more 
specialized organ, adapted only for climbing, and endowed in 
different plants with very various and some highly remark- 
able powers. To this subject Mr. Darwin has devoted more 
than half of his essay. An analysis of it must be deferred, 
for want of space. 

Near the close of the essay, under Hook-climbers, Mr. 
Darwin remarks that : — 

" Even some of the climbing Roses will ascend the walls of 
a tall house, if covered with a trellis ; how this is effected I 
know not; for the young shoots of one such Rose, when 
placed in a pot in a window, bent irregularly toward the light 
during the day and from it during the night, like any other 
plant ; so that it is not easy to understand how the shoots can 
get under a trellis close to a wall." 

Now we have had occasion to observe that the strong sum- 
mer-shoots of the Michigan Rose {Rosa setigcra, Mx., R. rubi- 
folia, R. Br.), trained on a latticed wall, are strongly disposed 
to push into dark crevices and away from the light ; they 



172 REVIEWS. 

would, many of them, pretty surely place themselves under 
the trellis, and the lateral shoots of the next spring would 
emerge as they seek the light. We suspect this is also true 
of the Sweet Brier. 

Twiners and Leaf-climbers having been considered, Ten- 
dril-bearers, which are the highest style of climbing plants, 
next demand our attention. But our analysis of this im- 
portant part of Mr. Darwin's treatise must be dispropor- 
tion ably brief. 

There are two kinds of movement exhibited by plants, 
which should be distinguished. 1st, Automatic, usually con- 
tinued movements, not set in action by extraneous invitation. 
The gyratory movement of the small leaflets of Desmodium 
gyrans is an exalted instance of this. 2d, Movements in con- 
sequence of the contact or action of an extraneous body, — of 
which those of the leaves of the Sensitive Plant may be taken 
as the type. Twining stems, as has been seen, strikingly ex- 
hibit the first, and their coiling around a support is a conse- 
quence of it. 

Tendrils for the most part execute both kinds of move- 
ment. They revolve, with some exceptions, like twining 
stems ; and they are all more or less sensitive to contact, — 
usually more so than the petioles of leaf -climbers, — bending 
toward the impinging body so as to hook or clasp around it, 
if the size will allow. Different tendrils act differently in 
some respects, some revolving freely, and sweeping wide 
circuits, some less evidently, and some, like those of the Vir- 
ginia Creeper, do not revolve at all, but turn from the light 
to the dark. But whether the tendril is the homologue of 
a leaf, or of a stem (or of a peduncle, which is the same 
thing), appears to make no difference in its action. On the 
other hand their diversity of gifts in one and the same fam- 
ily, or even in species of the same genus, is very remark- 
able, as may be seen especially in the Bignonia Family, the 
Grape Family, etc. So, also, the tendrils are commonly aided 
in their endeavors by the revolving of the internodes of the 
stem, but sometimes not, even in plants of the same genus 
or family. Mr. Darwin takes up tendril-bearing plants by 



HABITS OF CLIMBING PLANTS. 173 

natural families, beginning with Bignoniacece, which order 
contains tendril-bearers, leaf-climbers, twiners, root-climbers, 
and various combinations of these diverse modes. We, how- 
ever, will first consider the tendrils of the Gourd, and Pas- 
sion-flower families, regarding them as typical and simple 
representatives of tendril-climbers. 

Passiflora gracilis, a delicate annual species, lately in- 
troduced into the gardens, of the easiest cultivation, one 
which differs from most of its relatives in the young inter- 
nodes having the power of revolving , is said by Mr. Dar- 
win to exceed all other climbing plants in the rapidity of 
its movements, and all tendril-bearers in the sensitiveness of 
its tendrils. In the latter respect it decidedly surpasses our 
Echinocystis ; but it is nearly if not quite equalled by Sicyos, 
in which the coiling upon contact was first noticed as a vis- 
ible movement. The revolving internodes, when in the best 
condition, make almost hourly revolutions, and the long, deli- 
cate, straight tendrils revolve nearly in the same manner and 
at the same rate. The sensitiveness of the tendril, when full- 
grown, is correspondingly great, a single light touch on the 
concave surface of the tip causing a considerable curvature. 
" A loop of soft thread weighing -£. 2 d of a grain, placed most 
gently on the tip, thrice plainly caused it to curve, as twice 
did a bent bit of thin platinum wire weighing ^th of a grain ; 
but this latter weight, when left suspended, did not suffice to 
cause permanent curvature." After touch with the twig, the 
tip begins to bend in from 25 to 39 seconds. After coiling 
into an open helix upon transient irritation, they soon 
straighten again, recovering their sensibility ; but if left in 
contact, the action continues. We found it a pretty experi- 
ment, last summer, during the warmest days, to bring the upper 
part of an outstretched tendril by its inner or concave side 
against a twig or cord, and to see how promptly it would clasp 
it, revolving its free apex round and round it. A curious dis- 
crimination in the sensibility of such tendrils is mentioned 
by Mr. Darwin, as follows : — 

" I repeated the experiment made on the Echinocystis, and 
placed several plants of this Passiflora so close together that 



17-1 REVIEWS. 

the tendrils were repeatedly drugged over each other ; but no 
curvature ensued. I likewise repeatedly flirted small drops 
of water from a brush on many tendrils, and syringed others 
so violently that the whole tendril was dashed about, but they 
never became curved. The impact from the drops of water 
on my hand was felt far more plainly than that from the 
loops of thread (weighing ^d of a grain) when allowed to 
fall upon it, and these loops, which caused the tendrils to be- 
come curved, had been placed most gently on them. Hence 
it is clear, either that the tendrils are habituated to the touch 
of other tendrils and to that of drops of rain, or that they are 
sensitive only to prolonged though excessively slight pressure. 
To show the difference of the kind of sensitiveness in different 
plants, and likewise to show the force of the syringe used, I 
may add that the lightest jet from it instantly caused the 
leaves of the Mimosa to close; whereas the loop of thread 
weighing gVd of a grain, when rolled into a ball and gently 
placed in the glands at the base of the leaflets of the Mimosa, 
caused no action." — (p. 90.) 

Of Cucurbitaceous tendrils, the most active, after those of 
Sicyos (which Mr. Darwin has not observed), are those of 
Echinocystis lobata. The internodes and tendrils revolve in 
about an hour and three quarters, the former sweeping a circle 
or ellipse of two or three inches in diameter, the latter often 
one of 15 or 16 inches in diameter. Perhaps the most re- 
markable appearance of discrimination in tendrils is that 
which Mr. Darwin first noticed in this plant, but which may 
be seen in others, — and which he thus describes : — 

"I repeatedly saw that the revolving tendril, though in- 
clined during the greater part of its course at an angle of 
about 45° (in one case of only 37°) above the horizon, in one 
part of its course stiffened and straightened itself from tip to 
base, and became nearly or quite vertical. . . . The tendril 
forms a very acute angle with the extremity of the shoot, 
which projects above the point where the tendril arises ; and 
the stiffening always occurred as the tendril approached and 
had to pass, in its revolving course, the point of difficulty, — 
that is, the projecting extremity of the shoot. Unless the ten- 



HABITS OF CLIMBING PLANTS. 175 

dril had the power of thus acting, it would strike against the 
extremity of the shoot, and be arrested by it. As soon as all 
these branches of the tendrils begin to stiffen themselves in 
this remarkable manner, as if by a process of turgescence, and 
to rise from an inclined into a vertical position, the revolving 
movement becomes more rapid ; and as soon as the tendril 
has succeeded in passing the extremity of the shoot, its re- 
volving motion, coinciding with that from gravity, often 
causes it to fall into its previous inclined position so quickly, 
that the end of the tendril could distinctly be seen travelling 
like the minute-hand of a gigantic clock." — (p. 75.) 

Cucurbitaceous tendrils are mostly compound, in this case 
three-forked. When one of the lateral branches has firmly 
clasped any object, the middle branch continues to revolve. 
If a full-grosvn tendril fails to reach and lay hold of any 
object, it soon ceases to revolve, bends downwards, and coils 
up spirally from the apex. Indeed it often coils while still 
outstretched and revolving, the tendency to shorten (as we 
presume) on the inner side from the tip downward, which is 
usually brought into action by contact with an extraneous 
body, at length operating spontaneously. Uncaught tendrils 
when they thus coil up throw themselves of course into a simple 
helix or spire. One end being free, this is the simple and 
necessary consequence of the relative shortening of the con- 
cave side, sufficiently continued. 

In a caught tendril, the relative shortening of one side, 
(through which the tip hooks round and fixes itself to the sup- 
porting object), being propagated downwards, the whole now 
throws itself into a spiral form, — with more or less prompti- 
tude according to the species, — thus pulling the free portion 
of the tendril-bearing shoot nearer to the support, and within 
easier reach of the next tendril above. Both ends of the ten- 
dril being fixed, and the winding round an axis (real or imag- 
inary) necessarily involving or being a twist, it is certain that 
the caught tendril cannot now coil into a simple spiral, but 
that the spire will at least be double, a coil near one end of 
the tendril in one direction requiring the other to twist in the 
opposite direction, unless indeed it undergoes torsion. So, as 



176 REVIEWS. 

is familiarly known, there is at least one neutral point in 
a caught and coiled -up tendril, usually in the middle, the 
turns on one side of it running from right to left, on the other 
side from left to right. That the coils, whether simple or 
double and reversed (as the case may be), are not determined 
by any peculiarity in the tendril, but merely by the relative 
shortening of one side, may be readily shown by a thread cut 
from a piece of india-rubber, of unequal tension of the two 
sides ; this, when stretched and allowed to shorten while the 
two ends are held fast in the same plane, forms at once a pair 
of reverse coils, or three or four such coils, just as caught 
tendrils do. 

Mr. Darwin explains the point by analogous practical illus- 
trations, lie shows, moreover, that an important service ren- 
dered by the coiling or spiral contraction " is that the tendrils 
are thus made highly elastic." In the Virginia Creeper, where 
the ends of the compound tendrils are peculiarly attached, 
" the strain is thus equally distributed to the several attached 
branches of a branched tendril ; and this must render the 
whole tendril far stronger, as branch after branch cannot sep- 
arately break. It is this elasticity which saves both simple 
and branched tendrils from being torn away during stormy 
weather. I have more than once gone on purpose, during 
a gale, to watch a Bryony growing in an exposed hedge, with 
its tendrils attached to the surrounding bushes ; and as the 
thick or thin branches were tossed to and fro by the wind, the 
attached tendrils, had they not been excessively elastic, would 
have been instantly torn off and the plant thrown prostrate. 
But as it was, the Bryony safely rode out the gale, like a ship 
with two anchors down and a long range of cable ahead, to 
serve as a spring as she surges to the storm." 

Moreover, while unattached tendrils soon shrink up or 
wither and fall off, as we observe in the Grapevine, Virginia 
Creeper, etc., these same plants show how an attached tendril 
thickens and hardens, gaining wonderfully in strength and 
durability. In a Virginia Creeper, " one single lateral branch- 
let of a (dead) tendril, estimated to be at least ten years old, 
was still elastic and supported a weight of exactly two pounds. 



HABITS OF CLIMBING PLANTS. 177 

This tendril had five disk-bearing branches, of equal thickness 
and apparently equal strength ; so that this one tendril, after 
having been exposed during ten years to the weather, would 
have resisted a strain of ten pounds." 

Our space will not allow even an abstract of Darwin's 
account of the admirable adaptations and curious behavior of 
various tendrils, even of some very common plants ; as for 
instance of the familiar Cobcea scanclens, in which (the stem 
and the petioles being motionless) the great compound tendril 
borne at the summit of the leaf executes large circular sweeps 
with remarkable rapidity, carrying round an elaborate flexible 
grapnel, consisting of its five subdivisions, from 50 to 100 
in number, which are very sensitive even to a slight touch, 
bending in a few minutes toward the touched side, so that 
they clasp twigs very promptly, and all tipped with minute, 
double or sometimes single, sharp hooks, which catch in little 
inequalities, and may prevent the tendril branchlets from 
being dragged away by the rapid revolving movement before 
their irritability has time to act, while the still free ones pro- 
ceed to arrange themselves, by various queer and complicated 
movements, so as to secure the most advantageous hold ; then 
contracting spirally so as to bring other portions up within 
reach of the support, until all are inextricably knotted and 
fastened, and finally growing stouter, rigid and strong, bind- 
ing the plant firmly to its support. 

We cannot omit all mention of Big no?} i 'a capreolata, a not 
uncommon climber of our Southern States, of which we espe- 
cially wish to obtain fresh seeds or young plants, that we may 
ourselves observe the remarkable behavior of its tendrils which 
Mr. Darwin describes. These are said to turn from the light, 
as in many other cases ; they will clasp smooth sticks, but soon 
lose their hold and straighten themselves again. A rough, 
fissured, or porous surface alone satisfies them ; their young 
tips seek and crawl into dark holes and crevices, in the man- 
ner of roots ; then they develop their hooked extremity, and, 
especially when they meet with any fibrous matter, the hook 
swells into irregular balls of cellular tissue, which first adhere 
to the fibres by a viscid cement, and then grow so as to de- 



178 REVIEWS. 

velop thorn. This tendril can do nothing with a smooth post, 
fails to attach itself to a brick wall, but is well adapted to 
climb trees with rough and mossy bark. 

The Virginia Creeper also turns its tendrils from the light, 
and, although they will occasionally clasp a slender support, 
iu the manlier of its relative the Grapevine, they uniformly 
seek dark crevices, or especially broad flat surfaces, as a wall, 
a rock, or the trunk of a tree. Having brought their curved 
tips into contact with such a surface, these swell and form, in 
the course of a few days, the well-known disks or cushions by 
which they firmly adhere. Here is a tendril-climber which 
emulates a root-climber, such as the Ivy, in the facility with 
which it ascends smooth trunks, rocks, or walls. 

A very short chapter is devoted to Hook-climbers and Root- 
climbers. The stems of the latter are said to " have usually 
no power of movement, not even from the light to the dark. 
But Iloya carnosa, which twines, also climbs by rootlets 
spreading over the face of a damp wall ; and Tecoma radi- 
oing (our Trumpet Creeper) exhibits in its young shoots 
some vestiges of the revolving power with which its twining 
relatives are endowed." 

In a dozen pages of Concluding Remarks, Mr. Darwin 
gives much interesting matter in the way of deduction and 
speculation, which it would be difficult to condense into an 
abstract. 

Plants become climbers, he remarks, in order to reach the 
light, aud expose a large surface of leaves to its action and 
that of the free air. Their advantage is, that they do this 
with wonderfully little expenditure of organized matter in 
comparison with trees, which have to support a heavy load 
of branches by a massive trunk. Of the different sorts of 
climbers hook-climbers are the least efficient, at least in tem- 
perate countries, as they climb only in the midst of an en- 
tangled mass of vegetation. Next root-climbers, which are 
admirably adapted to ascend naked faces of rock ; but when 
they climb trees they must keep much in the shade, and fol- 
low the trunk ; for their rootlets, can adhere only by long- 
continued and close contact with a steady surface. Thirdly, 



HABITS OF CLIMBING PLANTS. 179 

spiral twiners, with leaf-climbers and tendril-bearers, which 
agree in their power of spontaneously revolving and of grasp- 
ing objects which they reach, are the most numerous in kinds, 
and most perfect in mechanism ; they can easily pass from 
branch to branch, and securely ramble over a wide and sun- 
lit surface. 

After adducing some considerations in support of his 
opinion that both leaf-climbers and tendril-bearers " were 
primordially twiners, that is, are the descendants of plants 
having this power and habit," Mr. Darwin asks : " Why 
have nearly all the plants in so many aboriginally twining 
groups been converted into leaf-climbers or tendril-bearers ? 
Of what advantage could this have been to them ? Why did 
they not remain simple twiners ? We can see several rea- 
sons. It might be an advantage to a plant to acquire a 
thicker stem, with shoi-t internodes, bearing many or large 
leaves; and such stems as are ill fitted for twining. Any 
one who will look during windy weather at twining plants 
w r ill see that they are easily blown from their support ; not 
so with tendril-bearers or leaf-climbers, for they quickly and 
firmly grasp their support by a much more efficient kind of 
movement. In those plants which still twine, but at the same 
time possess tendrils or sensitive petioles, as some species of 
Bignonia, Clematis, and Tropaeolum, we can readily observe 
how incomparably more securely they grasp an upright stick 
than do simple twiners. From possessing the power of move- 
ment on contact, a tendril can be made very long and thin ; so 
that little organic matter is expended in their development, 
and yet a wide circle is swept. Tendril-bearers can, from 
their first growth, ascend along the outer branches of any 
neighboring bush, and thus always keep in the full light ; 
twiners, on the contrary, are best fitted to ascend bare stems, 
and generally have to start in the shade. . . . 

" The object of all climbing plants is to reach the light and 
free air with as little expenditure of organic matter as pos- 
sible ; now, with spirally-ascending plants the stem is much 
longer than is absolutely necessary ; for instance, I measured 
the stem of a Kidney-bean which had ascended exactly two 



180 REVIEWS. 

feet in height, and it was three feet in length. The stem of a 
Pea, ascending by its tendrils would, on the other hand, have 
been but little longer than the height gained. That this sav- 
ing of stein is really an advantage to climbing plants I infer 
from observing that those that still twine, but are aided by 
clasping petioles or tendrils, generally make more open spires 
than those made by simple twiners." — (p. 110.) 

The gradations between one organ and another, and their 
special endowments, and the great diversity of their move- 
ments, are illustrated at length ; and the very large number 
of natural families which exhibit these endowments, in some 
of their members, is indicated ; and it is noted that two or 
three genera alone have those powers in some of the largest 
and best denned natural orders, such as Comjjositce, Ru- 
biacece, Liliacece, Ferns, etc. ; from which he infers " that 
the capacity for acquiring the revolving power, on which 
most climbers depend, is inherent, though undeveloped, in 
almost every plant in the vegetable kingdom " (p. 117). 

Mr. Darwin somewhere throws out the remark that the 
larger number, and the most perfectly organized climbing 
plants, as of the scandent animals, belong to one country, 
tropical America. 

In abruptly closing these extracts and brief commentaries, 
we would add, that the Linngean Society has issued a sepa- 
rate reprint of this charming treatise, thus opening to it a 
wider circle of readers. 



WATSON'S BOTANY OF THE 40TH PARALLEL. 

AVe propose to notice this volume x particularly : indeed it 
well deserves a more thorough examination and more ex- 
tended review than our time and space will now allow us to 
devote to it. It is published "by order of the Secretary of 
War, under the authority of Congress," as one of the Engi- 

1 United States Geological Survey of the Jfith Parallel. V. Botany. By 
Sereno Watson. Washington, 1871. (American Journal of Science and 
Arts, 3 ser., iii. pp. G2 and 148.) 



BOTANY OF THE 40TH PARALLEL. 181 

neer Corps series, has been carefully edited and beautifully 
printed, so that the volume is every way an attractive one. 
Errors of the press are to be found, but they are apparently 
few, and the whole typography is remarkably excellent for the 
Government Printing Office. Our comparison is naturally 
with corresponding volumes of the Pacific Railroad Survey, 
and of the Mexican Boundary Survey, upon which the pres- 
ent volume is a notable improvement. The forty plates, filled 
with well chosen subjects, if not of the very highest style, 
are so well done and of such excellent promise that the name 
of the draughtsman (who is new to this class of work, we 
believe), Mr. J. H. Emerton, of Salem, Massachusetts, should 
properly have been appended to them. 

The General Report, of 53 pages, forms a separately 
paged introduction to the " Catalogue," as it is termed with 
excessive modesty, i. e., the systematic account of the plants 
collected, which makes up the principal bulk of the volume. 
This General Report will naturally be most interesting to 
general readers and naturalists, but no less so to special bota- 
nists. It is thoroughly readable matter, and we expect to 
see it reproduced in the scientific journals. Four or five 
pages sketch the geographical features of the region, tersely 
and clearly. But, when a stream of water is said to " become 
demoralized with alkali and is lost," we could wish that this 
popularized use of the word were buried with it. The mete- 
orological notes, with tabulated observations by thermometer, 
evaporator, etc., are equally interesting, displaying the dry- 
ness of the Great Basin, its cold winters and hot summers. 
The notes on the general character of the vegetation picture 
to us the botanical aspect of the region, the relative preva- 
lence of the predominant species, the slow and cross-grained 
growth of what timber there is in the canons, etc. A dead 
branch, apparently of Pinus monophylla, 8 inches in diame- 
ter, had the fibres so twisted that in 7 feet they made four 
complete circuits. A saw-mill in Ruby Valley offered the 
opportunity of ascertaining the age and dimensions of several 
specimens of Pinus Jlexilis from the upper canons of the 
Humboldt Mountains ; sections from 22 to 30 inches in 



182 REVIEWS. 

diameter showed from 400 to 48G annual rings. The " ever- 
lasting sage-bush," Artemisia tridentata, displayed 65 rings 
on a section 8 inches in diameter, 37 upon 4 inches, etc. A 
Juniperus occidentalism 12 inches in diameter, showed 250 
rings. Cercocarpus ledifolius, it appears, may form a trunk 
of 2 feet in diameter, with 1G0 rings. The alkaline species, 
aquatic and meadow species, those of the drier valleys and 
foot-hills, the mountain species, etc., are separately enumer- 
ated ; the introduced species, about 30 in number, are re- 
corded, and finally the number of indigenous genera and 
species is given under their orders, and their distribution in 
the basin or over the borders on either side is tabulated. Of 
the 1141 species of the basin and of the Wahsatch and 
Uintas, GO per cent, appear to inhabit also the Pacific slope, 
about 60 per cent, are not found east of the Eocky Moun- 
tains, 15 per cent, only approach the Mississippi or the Sas- 
katchewan, 25 per cent, approach the Atlantic, 17 per cent, 
are Mexican or southern, and nearly 15 per cent, are Arctic. 

A few pages at the close are devoted to the consideration of 
the agricultural resources of the basin, the limit to which is 
fixed by the deficiency of water. " The most fertile localities 
lie at the base of the Sierras ; but, as a rule, there is an appar- 
ent absence everywhere of a true soil or mould resulting from 
the decomposition of vegetable matter." A moderate amount 
of alkali in the soil appears not to be detrimental to culture. 
The soil which produces " sage bush " seems to be alwaj^s cul- 
tivable when it can be irrigated. With the present supply of 
water, most economically used, it is thought that only one 
thousand out of 34,000 square miles of northern Nevada could 
be brought under cultivation ; of the southerly portion and 
of western Utah much less. Eastern Utah, with larger and 
more constant supplies of water from the Wahsatch and the 
Uinta Mountains, is much more favorably situated. The 
absence of graminivorous animals, excepting rabbits in the 
valleys and rarely a few mountain sheep or antelopes in the 
higher ranges, shows that the country is ill adapted for graz- 
ing. Eurotia lanata and a few other Chenopodiaceous plants 
are eaten by sheep as a substitute for grass. 



BOTANY OF THE 40TH PARALLEL. 183 

Mr. Watson raises the question whether — considering the 
amount of low shrubby and perennial vegetation which inhabits 
the plains and thrives without irrigation — these plants them- 
selves, or some more serviceable substitutes equally adapted to 
the climatic conditions, may not be turned to some profitable 
account under the necessities of a future population ; and 
whether, in time to come, some forms of orchard, vineyard, or 
tree-culture may not possibly be made to thrive in that region. 
He finds that the present plants on the whole are not lacking 
in expansion of foliage or succulence, at least that the more 
prevalent plants had an average of from 55 to 80 per cent. 
of foliage or working surface ; and a series of rough but seem- 
ingly well-devised experiments demonstrated that they give 
off by evaporation daily an amount equal to three eighths of 
the weight of their available material. Dry as the soil appears 
to be, it is this, and not the atmosphere, that must furnish the 
supply to make good this loss. Yet water is rarely to be had 
under a depth of 100 to 300 feet, often not even at that depth. 
The porous soil must allow of the free upward diffusion of 
moisture, also of deep penetration of the roots from above. 

An excellent map is given, exhibiting the district from 
above the 42d parallel to below the 39th, on which the routes 
of the three several years are traced in colored lines, and the 
mountain ranges with the general configuration of the surface 
represented. We will endeavor hereafter to review the sys- 
tematic part of this work. 

(Second Notice.) — Under the modest name of a catalogue 
of the known plants of Nevada and Utah, Mr. Watson has 
given us a treatise, not to say a Flora, of a wide stretch of 
country between the Sierra Nevada and the Rocky Mountains, 
which is invaluable to the botanist studying the plants of that 
region in herbaria, and still more to explorers on the ground, 
— of which we hope there may be many. For not only are 
new or revised species described, but all species not contained 
in the common eastern manuals, etc., which every collector is 
supposed to possess ; the characters of western genera are 
appended in foot-notes, and synopses of recently elaborated 
genera — some of them reprints or translations of scattered 



184 REVIEWS. 

papers with corrections or additions, others original revisions 
by the editor himself — are added in an appendix, so as to 
afford every possible help to the student or collector who has 
not access to a full botauical library, and indeed most accept- 
able facilities to those few who have. 

Alter thus calling attention to a volume of so much im- 
portance, we propose to restrict our comments to sundry 
details of criticism, or points of information, where opportu- 
nity occurs. 

Under Thalictrum Fcnclleri some synonyms are adduced 
which are not all certain ; as there is another Oregon species 
which has been confounded with T. dioicum, but is distinct 
from both in the fruit, which was sparingly collected in the 
British Boundary expedition, and lately by Mr. Hall. 

Ranunculus alismcefolius var. montanus is essentially equiv- 
alent to the variety alismellus Gray ; although the speci- 
mens from the " head of Provo River in the Uintas " are a 
stouter and larger-flowered form, identical with Parrv's No. 
79, which we had wrongly named when distributed, and which 
may be rightly characterized as merely a dwarf mountain state 
of Geyer's Ii. alismcefolius. We may now add that there is 
a much older name for this species, especially for this moun- 
tain form of it, namely, It. Pseudo-ffirculus of Schrank, 
1842, a Songarian plant. It may also be noted that, while 
this species in eastern America takes the place of the Euro- 
pean It. Flammula, both occur on the western side of the 
continent (as also in Siberia), and in forms so much alike 
that only the character of the style and that of the petal and 
its scale (so well indicated by Mr. Watson) will serve to dis- 
tinguish them. Of amply developed It. Flammula — as large 
as any European form — copious specimens have been col- 
lected in Oregon last year, by Mr. Elihu Hall, and are soon 
to be distributed. 

As to R.fascicularis, there is no clear evidence that this 
species extends to California, Nevada, or even to Oregon. 
The plant referred to and so named in Ly all's collection, 
though not in fruit, is apparently R. orthorhynchus, a plant 
most rare in collections, but now, thanks to E. Hall's collec- 



BOTANY OF THE JfiTR PARALLEL. 185 

tion of last summer, likely to be supplied to botanists. As 
to Mr. "Watson's R. orthorhynchus var. alpinus, that is cer- 
tainly not of this species, but a wholly new one, unless it be the 
rare, and to us obscure, M. pedatijidus of Smith, or at least of 
Hooker. For since Schlechtendal's plant of that name has 
been referred to R. offinis, the Siberian one of Smith may also 
be of that species. 

A yellow-flowered Aquilegia, with flowers rather smaller 
and sometimes much smaller than those of A. Canadensis, and 
with spurs shorter than the widely spreading sepals, after the 
manner of A. formosa, and more or less curved (thus approach- 
ing the European type), which has been collected by Lyall, 
Bourgeau, and others, is now characterized as a new species, 
under the name of A.jlavesccns, Watson. It should be noted 
that this has been cultivated in European gardens, from seeds 
collected by Roezl, under the name of A. aurea, but it is 
doubtful if it is yet published under that name. 

The Cruciferce constitute an important order in the interior 
basin and its borders. One of Mr. "Watson's most notable 
discoveries is that of Brown's Parrya macrocarpa, hitherto 
found only on the Arctic coast. It was detected on the high- 
est peak of the Uintas, at an altitude of 12,000 feet. The 
next point of interest is found in our author's discoveries and 
views of plants of the Streptanthus and Thelypodium type. 
Two or three well-marked new species are introduced, and 
Xuttall's obscure Streptanthus cordatus is confidently identi- 
fied. In the present view this is the only Streptanthus of the 
collection ; Mr. Watson, having ascertained that several spe- 
cies, such as S. procerus, the curious S. crassieaidis, and two 
new species, have oblong seeds in a terete elongated pod, 
and cotyledons inclining to be incumbent in the manner of 
Thelypodium, combines them into his new genus, Caulanthus. 
And Iodanthus, with a few other species, some of which had 
already been excluded from Streptanthus, are referred, as had 
also been tentatively suggested, to Thelypodium. Which is 
all to be highly approved, except, perhaps, the expediency of 
the new genus, when all could be disposed in the two genera : 
Streptanthus for the species with flat or flattish pods, flat 



186 REVIEWS. 

seeds, and truly accumbent cotyledons ; Thelypodium, for 
those with more or less terete pods, narrow seeds, and more 
or less incumbent cotyledons. 

We are bound, moreover, to take steps for the suppression 
of a nominal species which is here introduced in consequence 
of our own short-sightedness. In an evil moment we gave the 
name of Smeloicskia ? Calif ornka to a plant of Professor 
Brewer's collections, thought to be perennial, with exceedingly 
short few-seeded pods. This Mr. Watson identified with a 
common Sisymbrium of the region, distinguished from S. 
canescens by its seeds, strictly in a single series, and transfer- 
ring the name, calls it S. Californicum. He had overlooked 
an article in this Journal (for September, 1866) upon this 
Sophia group of Sisymbrium, from which it would have been 
seen that the plant in question is Sisymbrium incisum of 
Engelmann, and the later S. longipedicellatum of Fournier, 
besides one or two other names of the same author more or 
less strictly referable to it. 



DECAISNE'S MONOGRAPH OF THE GENUS PYRUS. 1 

A volume of Decaisne's great work — or rather of one of 
his great works — " Le Jardin Fruitier du Museum, un Icono- 
graphie de touts les Especes et Varietes d'Arbres Fruitiers 
cultives dans cet Etablissement," etc. (produced in first-rate 
style by Firmin Didot Freres), devoted to the genus Pyrus, 
is now before us. It is a complete monograph of the species 
of this genus, taken in its restricted sense, illustrated by 
figures of the wild types, and also of the cultivated races of 
those cider-pears known in France under the name of Sanger. 
There is a list of the cider-pears cultivated in the different 
provinces of France, a general alphabetical catalogue of all 
the published varieties of pears, and a table in which the syn- 
onyms are referred to the names severally adopted. The 
other volumes, and the illustrations of the edible varieties of 
1 American Journal of Science and Arts, 3 ser., iv. 489 ; x. 481. 



MONOGRAPH OF THE GENUS PYRUS. 187 

pears, may have more interest for the horticulturist. But the 
present attracts the special attention of the scientific botanist. 

As stated in the Introduction, Professor Decaisne entered 
upon his great undertaking more than twenty years ago, when, 
in the year 1850, he became the Professor of Culture. He 
cites the instructions under which the separate collection of 
fruit-trees was constituted, and the professor of culture was 
charged with its management, and was directed to bring to- 
gether all the known varieties, with all their names, " afin 
d'etablir une uniformite de nomenclature necessaire pour 
toutes les parties de la Republique." This is a decree of the 
National Convention, June 10, 1793. The collection which 
Decaisne has so diligently and acutely studied actually dates 
from the year 1792, when the fruit-garden of the Chartreux 
of Paris was broken up, and two trees of each variety trans- 
ported to the Jardin des Plantes. In 1793 it contained 185 
varieties. In 1824, when Thouin died, there were in it 265 
varieties of pears alone ; it has now more than 1400 varieties 
of this fruit. It is interesting and important to know that 
the collection still preserves the greater portion of the very 
types described a century ago by Duhamel. For seven years 
Professor Decaisne studied the incomparable collection under 
his charge, making drawings and analyses, in which he is so 
skilful, and an herbarium of their flowers and foliage, before 
he commenced the publication of the " Jardin Fruitier du 
Museum," which he is now bringing to a close. 

As to giving a correct nomenclature and available charac- 
ters, this is difficult enough, as all botanists know, for the 
species themselves (which must needs have, or be assumed to 
have, real distinctions) in any large genus, such as Quercus, 
Rosa, Rubus, and the like ; how much more difficult, even to 
impossibility, it must be in the case of cultivated varieties, of 
ever increasing numbers, usually named without system, some- 
times of mixed origin, and often too like each other to be dis- 
tinguished by any available descriptions. Here colored plates 
are a necessity ; and those of this great standard work, upon 
which no pains have been spared, leave little to be desired 
that art can supply. 



188 REVIEWS. 

In France alone they count about 800 sorts of pears ; the 
origin of most of them is unknown, and many are undoubtedly 
very ancient. Indeed, according to Jordan and his school 
these differences are primitive, and the so-called races and 
varieties, both of wild and cultivated plants, represent so many 
closely related species. But M. Decaisne, not content with 
the rcductio ad absurdum of having about 2000 species of 
pears to be dealt with, proceeded to an experimental demon- 
stration of the variability of the cultivated races. He sowed 
the seeds from four very distinct varieties in 1853, the Poire 
d'Angleterre, the Bosc, the Belle Alliance, and the Cirole. 
Of the last the four trees raised bore fruit of four different 
forms. From the Belle Alliance he obtained, in this first 
generation, nine new varieties, none of them representing the 
parent, neither in the form, size, color, nor even the time of 
ripening of the fruit. The Bosc equally produced new varie- 
ties. Of the Angleterre nine trees produced as many new 
forms, one of them a winter-pear similar to the Saint Ger- 
main, another apple-shaped fruit identical with one which 
was raised from the Belle Alliance. On plate 33, Decaisne 
gives figures of six different pears raised from the Angleterre. 
These results even led him to doubt the cases cited by Darwin 
of the reproduction of certain pears from seed. He insists, 
moi'eover, that very bad fruits may be raised from choice cul- 
tivated pears, and that good varieties may be obtained from 
the seeds of wild pears. The latter is not what one would 
expect in the first generation. 

Our author proceeds to state that the trees raised from seed 
taken from the same fruit differed, not merely in their fruits 
and in the time of ripening, but no less in their flowers and 
in the form of the leaves. Some were thorny, others thorn- 
less ; some produced slender shoots, others thick and stout 
shoots, etc. It is worth noticing, however, that no mention is 
made of any precautions to prevent cross-fertilization of the 
flowers from which the seeds planted were derived, which 
might have influenced the product through the now well-ascer- 
tained influence of the pollen upon the pericarp. We per- 
ceive, however, that he would regard this as unimportant, 



MONOGRAPH OF THE GENUS PYRUS. 189 

since pear varieties are of the lowest grade, incapable of pro- 
pagating fruit by close-fertilization, and therefore wholly un- 
likely to impress by their pollen any characteristic upon the 
pericarp of another variety. 1 A large part of the Introduc- 
tion is occupied with further evidence that the Pear-ti'ees of 
cultivation are all of one species, from which have proceeded 
six races, completely fertile inter se, and varieties ad infinitum. 
In this respect the Pear-tree has but followed the example of 
most fruit and fruit-trees, and of the Grains, etc., which had 
apparently diverged into races, or distinct but closely related 
types, in very early times, and those under cultivation have 
themselves varied and subdivided more and more. Finally, 
M. Decaisne maintains, seemingly with good reason, that to 
combine into one genus the Apple, Pear, Quince, Sorb, and 
Mountain Ash, as done by Linnaeus and followed by the latest 
authorities, is to misconceive the laws of the natural system ; 
that " to unite generically these plants, which differ in the 
character of their wood, the vernation of their leaves, their 
inflorescence, the aestivation of the corolla, and the structure 
of their fruit," logically leads to the combination of all Po- 
macecB into one genus. He accordingly restricts the genus 
Pyrus, or (restoring the classical orthography) Pirus, as did 
Tournefort and Jussieu, to the Pear proper. To the organo- 
graphy of this restricted genus, from the wood to the embryo, 
a full chapter is devoted. In the course of this the relative 
systematic value of characters observed is bi'ought out. He 

1 Yet the Apple, which is in the same ease, does so. An interesting 
instance of this kind lately came under our notice, an apple from a Spitz- 
enberg tree, one half (at least as to the surface) Spilzenberg, the other 
half Russel. A tree of the latter fruit stood ahout 200 yards off. Several 
cases of this sort are known, in which, as in this, the division is into two 
exactly equal parts of the circumference, and the line of demarcation 
abrupt. This is quite unexpected, as the Secretary of the Smithsonian 
Institution, who sent us the fruit, remarked ; for as the styles and carpels 
were five, we should have expected the division to be into fifths, and ac- 
cording to the number of the stigmas which were acted upon by the for- 
eign pollen. It is, moreover, to be noticed that the action of the pollen 
in this case is manifest upon what is morphologically the calyx, not upon 
the pericarp. The apple we refer to was grown in the orchard of William 
Wicksham, of Washington County, Pennsylvania. — A. G. 



190 REVIEWS. 

notes that the vernation of the leaves is involute in Pyrus, 
but not in Cydonia, Mespilus, and Aria ; that the cottony- 
leaved varieties, no less than the smooth ones, are glabrous 
in the seedling stage ; that all varieties of the common Pear 
blossom at Paris whenever, in the month of April, the mean 
temperature reaches about 10° Centigrade, without perceptible 
difference between the earliest and the latest-ripening varie- 
ties ; that the aestivation of the corolla is convolute in Cydo- 
nia, but imbricate in the Pear, although ordinarily quincuncial 
in other Pomacece (but in the two diagrams of Pear-flowers 
on Plate A, one has the quincuncial, i. e., in our view typically 
imbricative aestivation of the corolla ; in the other, there is 
only one wholly outer and one inner petal, — a combination of 
the quincuncial and the convolute modes which often occurs, 
but which need not be taken as the type of imbrication) ; that 
there are two types as to size of the corolla in the common 
Pear, the smaller flowered type comprehending most of the 
cultivated varieties ; that the odor of Pear-blossoms is rather 
disagreeable than otherwise, in contrast with those of Malus, 
which are sweet-scented. Moreover, the anthers in the Pear 
genus are tinged with violet ; those of the Apple genus are 
yellow. 

As to the morphology and development of the gynaecimn, 
Decaisne reproduces in full the note which he published in 
the " Bulletin of the Botanical Society of France " in 1857. 
From his investigation it appears that the five carpels in their 
early development are free and distinct in the concave centre 
of the flower ; that at a later stage, when the concave recepta- 
cle has become much deeper, a cellular tissue develops from 
its base and inner face, moulds itself around and over the car- 
pels, so as separately to envelop them, except at their inner 
angle, while it carries up the petals and stamens, and forms 
the perigynous disk upon which they are inserted ; this forms 
the core or central part of the flesh of the fruit, which we 
have always regarded as receptacle, never ceasing to protest 
against the still prevalent notion (continued in the latest gen- 
eral works), that the cartilaginous or bony "cells" are "en- 
docarp." But, while we were disposed to regard the whole 



MONOGRAPH OF THE GENUS PYRUS. 191 

exterior flesh as calyx, Professor Decaisne (no doubt cor- 
rectly) regards it as mainly receptacle or axis, an hypan- 
thium which in common pears is largely a hypertrophy of the 
peduncle, after the fashion of Anacardium. 

In the proper Pear genus, the ovules never exceed a single 
pair ; this should therefore enter into the generic character. 

" Theophrastus had already remarked that the older the 
Pear-tree, the more prolific, and every day's experience con- 
firms the justice of this observation." The gritty grains or 
lignified cells which are so abundant in the flesh of many 
sorts of pears are not wholly absent from any of them. To 
them is due the roughish surface of the skin, as contrasted 
with the smooth skin of apples. It is curious to remark that 
Meyen, in his " Pflanzen-Pathologie," considered the gritty 
grains to be a disease which attacked pears and quinces. 

It appears that pear-growers are able to produce fruits of 
abnormal size by supporting the growing pear from under- 
neath, instead of allowing it to hang on the peduncle. M. 
Decaisne has seen Poires de Livre of a kilogram, Glou- 
morceau of GOO grams, and a Chaumontel of 700 grams 
weight, produced in this way. 

The testa of all Pomaceous seeds is smooth and more or 
less mucilaginous, except of a Photinia, in which it is retic- 
ulated. The cotyledons are accumbent relative to the rhaphe, 
except in a Photinia, Cotoneaster, Pyracantha (Cratcegus 
jPyracantha, Pers.), and Eriobotrya, in which they are incum- 
bent. At first there is a thin layer of albumen, which disap- 
pears at maturity of the seed. 

Pears are commonly grafted upon a Quince stock. But it 
is confidently asserted, and generally supposed, that there are 
more than forty varieties which absolutely refuse this union, 
and which are therefore managed by subgrafting upon a Pear 
stock of a proper sort which has itself been engrafted upon 
the Quince. But, as Professor Decaisne remarks, horticul- 
turists are too apt to generalize their impressions and to limit 
nature to the narrow horizon of their own practice. Upon 
the first trial of the experiment under his own observation, he 
succeeded with twenty of these antipathetic varieties without 



192 REVIEWS. 

the least difficulty ; but some (among which are the Clairgeau 
and the Bosc) obstinately refuse to unite with the Quince 
stock. lie naturally discredits the assertion made by Cabanis 
and by Downing (cited by Darwin), that when certain pears 
are grafted on the Quince, their seeds produce trees of types 
different from those which they do when they are raised upon 
a Pear stock. Decaisne found, as already stated, that Pear- 
seeds produce indifferently new varieties in any case ; that 
these varieties are not at all fixed into races. He regards as 
wholly unproven all the assertions that the fruit is ameliorated 
or in any degree altered by grafting upon a Quince or any 
other stock. He records a very exceptional instance in which 
the antipathy of the Pear to the Apple as a stock was so far 
overcome that the graft survived at least six years, but with- 
out vigor, and bore fruit ; still this antipathy confirms the 
generic difference between Pyrus and Mains. 

We must pass over the sections on the diseases of the Pear, 
and the parasitic plants and insects hurtful to it : while as to 
that on the classification of the Pears of cultivation, we may 
mention merely the conclusion, which is, that a natural classi- 
fication of Pears is thus far an impossibility ; and that in 
practice nothing better can be done than to follow the exam- 
ple of the older pomologists, who arranged them according to 
the period of ripening. A general list of the adopted names 
of the published varieties of cultivated Pears, alphabetically 
arranged, fills four pages of the volume. A list of their syno- 
nyms, in which each is referred to the adopted name, fills over 
12 pages ! Then follows a list of pears classed according to 
the period of maturing, and in which the best varieties are 
designated. 

Finally comes a botanical monograph of the genus Pyrus, 
with a full generic character, and descriptions and figures of 
the races, as he would term them, considering as he does all 
known forms of the restricted genus as a single and very 
polymorphous species. 

The six races are : 1. The Celtic, Proles Armorica?ia, of 
three quasi-species, P. cordata, Poissieriana, and longipes. 
2. The Germanic, Proles Gcrmanica, or Pyrus communis, 



MONOGRAPH OF THE GENUS PYRUS. 193 

including our common pears, both pear-shaped and apple- 
shaped, " both forms being often met with upon the same 
tree." Under this head Professor Decaisne gives some inter- 
esting pages upon the history of the cultivation of pears in 
France, which cannot be ancient, and of cider (perry) as 
a drink. It appears that it took the place of beer in the 
north of France in the fifteenth century or later, and is now 
giving way to wine and perhaps beer again ; and that pears 
would have disappeared before this from a part of Normandy, 
were it not that they are carried in immense quantities to 
Epernay, where they are used in the manufacture of cham- 
pagne. 3. The Hellenic Race, which comprises P. parviflora 
and three other subspecies. 4. The Pontic Race, P. salici- 
folia and its allies. 5. The Indian Race, P. Pashia and 
its relatives. 6. The Mongolian Race, P. Sinensis and its 
varieties. As one turns over the excellent plates one can 
hardly be persuaded that such extremely diverse forms can 
practically be regarded as of one species. 

A list of the species remanded from Pyrus to other genera 
shows that the result of our author's prolonged and sagacious 
study is to increase the genera about as much as he diminishes 
the species of the Linnaean Pyrus. 

A detailed analysis of Decaisne's monograph of the genus 
Pyrus was given in this Journal (3 ser., iv. 489, Dec, 1872). 
Some of the views taken in that work are fully expounded in 
the present paper, : which embracing the results of a prolonged 
study of an important group, by a botanist of great experience 
and ability, is worthy of particular attention. As the veteran 
author states it : — 

" My principal object is here to call the attention of bota- 
nists to certain characters which have been neglected in sys- 
tematic works, by the aid of which the ancient genera merged 
in Pyrus by most of the recent systematists may be neatly 
circumscribed. Such is the constancy and the value of these 
characters that the details of organization peculiar to each 
generic group may be expressed by very general propositions, 

1 Memoire sur la Famille des Pomace'es. Par J. Decaisne (Nouvelles 
Archives du Museum, x. pp. 113, 192). Paris, 1ST5. 



194 REVIEWS. 

which is the very object of a good method. Indeed, when a 
special organization is common to a large number of different 
plants, it is evident that comparatively slight but constant 
modifications of this structure ought to be particularly attended 
to ; and this proposition seems to be especially true of the 
Pomacece." M. Decaisne puts foremost his strongest point 
when he declares of the Quince, that " the nature of its bark 
and wood, its prefoliation, inflorescence, the aestivation of the 
corolla, the structure of the ovary and of the fruit differ es- 
sentially from that of the Pears, among which certain bota- 
nists still class it." Rather than combine the Quince and the 
Japan Quince with Pyrus, we are confident that botanists 
will generally accept his Docynia, along with Chgenomeles 
Lindl. and Cydonia, as independent genera. The same may 
be said of Mespilus ; and it must be allowed that the charac- 
ter which Kunth had noticed and which Decaisne has turned 
to account, that of the deformation of one of the ovules which 
becomes a kind of stipitate hood for the other, being common 
to it and to Crataegus, indicates a relationship to the latter 
genus rather than to Pyrus. Much nicer and more question- 
able characters are assigned to the genera here re-established 
from Pyrus in the Candollean sense, to which we are in this 
generation accustomed. These are, Aronia, Pers., our Choke- 
berry (in which eight species are set up from what we take to 
be a single polymorphous one) ; Sorbus, Tourn., the Mountain 
Ash (the synonym S. mkrorarpa omitted from S. Americana, 
and S. sambucifolia is still taken to belong only to our west- 
ern coast, whereas it extends across the continent) ; Aria, 
Host., the Beam Trees, all of the Old World ; Torminaria, 
Eoem., for Pyrus torminalis ; also Cormus, Spach, for Sor- 
bus (7omestica,~L., the Service-tree of Europe, with Pyrus 
tri/obata, DC, and an allied species ; Micromeles, a new genus 
for four Himalayan species thus far little known ; lastly, 
Malus, Tourn. Here it is to be observed that M. diversifolia 
is held to be distinct from M. rivularis ; and that a subgenus, 
Chloromeles, proposed for M. an gust 'folia, our narrow-leaved 
Crab-Apple, thus widely separated from 31. coronaria, on 
account, as is stated, of its reddish anthers and the structure 



MONOGRAPH OF THE GENUS PYRUS. 195 

of the disk. Pyrus, Tourn., is thus brought down to the 
Pear ; and this, as Decaisne had formerly announced, to a 
single collective species, of six geographical proles or forms. 
We continue to write Pyrus from old habit and custom, not 
doubting, however, that Pirus is the correct orthography. 

Of Amelanchier, following Lindley, there are enumerated 
twelve species, six for the Old "World and six for North 
America, and there are names for four more. Without being- 
able to clear them up (and no wonder), Decaisne thinks that 
they may be distinguished into at least three groups, charac- 
terized by the distinct or united styles, and the glabrous or 
downy ovaries. We are continually impressed with the idea 
that there must be three or four American species, and the 
seeds may aid in their definition. But thus far the characters 
elude investigation. Peraphyllum, Xutt., referred by Ben- 
tham and Hooker to Amelanchier, has not been studied by 
Decaisne. When he examines the excellent specimens in 
flower and in fruit, which Mr. Siler has supplied from south- 
ern Utah, he will conclude that the genus must certainly be 
reinstated. The likeness is only in the peculiar structure of 
the fruit. 

As respects the remaining genera, the difference between 
this monograph and the disposition in Bentham and Hooker's 
"Genera Plantarum" is mainly this : Eriobotrya, with its bac- 
cate fruit (what is termed endocarp reduced to a soft pellicle), 
large turgid seeds with thickened cotyledons, and undulate 
petals, is upheld as a good genus ; Heteromeles is adopted 
from J. Rcenier for the Californian Photinia arbutifolia, 
and a second (probably not good) species, H. Fremontiana, 
is added. The characters appear to be the 10 instead of 20 
stamens, in pairs opposite the calyx-lobes, their filaments 
dilated at base and somewhat monadelphous. In the tabular 
conspectus the petals are said to have " prefloraison tordue," 
but in the jreneric character it is " sestivatione imbricativa vel 
convolutiva," the latter term with the French botanists mean- 
ing the same as imbricated, only more enrolling. The diagram 
represents the whole five petals with one edge covered, i. e., 
"tordue" or contorted (or, as we say, convolute), and so we 



196 REVIEWS. 

find them in all the flower-buds now examined. But before 
adopting the genus it may be well to examine the Photinice 
generally. Photinia, of which P. serrulata is the type, is 
characterized as having imbricative aestivation, and Decaisne's 
diagram represents it as regularly (i. c, quincuncially) so. 
But in P. prunifolia and in P. Blumei we find occasionally 
only one exterior petal, and the four others successively over- 
lapping in the " contorted " way ; and in one of AVallich's 
specimens of P. integrifolia the first flower-bud inspected 
showed the " contorted " aestivation complete. This is also 
the case in P. dubia (in one of Hooker's and Thompson's 
Khasyia specimens), and this Decaisne refers to Eriobotrya, 
which has imbricative aestivation. Next is Pourthiaea, a new 
genus of eleven Japanese and Indian species, the type being 
Photinia arguta, villosa, Icevis, etc., and the character, among 
others, "aestivatione contorta." But we as commonly find 
one petal wholly exterior. So we think it evident that the 
aestivation of the corolla furnishes no characters for the divi- 
sion of the genus Photinia. 

Finally, as to the proper stone-fruited genera, Pyracantha 
is adopted from J. Eoemer for Crataegus Pyracantha and 
an allied Indo-Chinese species, and placed near Cotoneaster ; 
and a character not before used is introduced, namely, the 
position of the cotyledons, which in this genus are, as regards 
the rhaphe, accumbent. 

There are eight plates, six of them filled with admirable 
dissections, neatly done upon stone by Riocreux from the 
author's sketches. 



ENGELMANN'S NOTES ON THE GENUS YUCCA. 1 

This modest title comprises the principal results of Dr. 
Engelmann's long study of a difficult genus of plants. Pur- 
suing his botanical investigations now for many years only 

1 Notes on the Genus Yucca. By George Engelmann, in Trans. St. Louis 
Acad. iii. St. Louis, 1873. (American Journal of Science and Arts, 3 ser., 
vi. 4G8.) 



NOTES ON THE GENUS YUCCA. 197 

in the intervals and spare moments of a busy and exacting 
professional life, Dr. Engelmann has made them tell most 
effectively and advantageously upon the science which num- 
bers him as a distinguished votary, by taking up one subject 
at a time and investigating it as thoroughly as possible. In 
this way he has mastered, in turn, our Cuscutce (upon which 
his earliest essay was published in this Journal, thirty-one 
years ago, and his latest was a full monograph of all the 
known species throughout the world), our Cactacece, our Mis- 
tletoes, Euphorbias, Junci, Callitriches, etc., not to speak of 
several other genera or groups, or taking account of his sedu- 
lous and long-continued study of our Oaks, and, above all, of 
our Coniferce. Nor need we look to this as the close of the 
series, but rather see before him " fresh fields and pastures 
new," and wish for him more time to expatiate in them. 
Upon the principle " to him that hath shall be given," he well 
deserves it, as having accomplished far more in these rescued 
moments than others who could mainly devote their days as 
well as nights to scientific work. Almost without exception 
these monographs relate to difficult subjects, and such as re- 
quire long-protracted investigation. This is also true of the 
present essay upon the genus Yucca. It is not a large one, 
only a dozen species being clearly made out ; but those of 
long cultivation in Europe have been much confused, and 
recent ones described without flowers, while fruit is rarely 
formed out of their native stations, and dried specimens of 
any completeness are difficult to make, so that means of com- 
parison are much restricted. 

The true anthesis, as is now shown, is nocturnal, the flowers 
remaining half closed during the day. The anthers, with 
comparatively large and few grains of glutinous pollen, open 
rather earlier than the flower. The tips of the style, which 
were naturally taken for stigmas, are now shown to be func- 
tionless, the stigmatic surface being the moist and glutinous 
lining of a stylar tube, which extends downward nearly to the 
cells of the ovary and even communicates directly with them. 
As soon as it became evident that fertilization must depend 
upon nocturnal insects, it was found that they were most fre- 



198 REVIEWS. 

quently and regularly visited by " white moths, which, usually 
in pairs, disported in the open flowers at dusk, and were found 
quietly ensconced in them when closed in the day-time." 
Professor Eiley of St. Louis, the distinguished entomologist, 
was at this point associated with Dr. Engelmann in the inves- 
tigation. The result has been given to the scientific world in 
his interesting memoir on Pronuba yuecasella, first read at 
the Dubuque meeting of the American Association for the 
Advancement of Science, in August, 1872, and now also pub- 
lished, as a sequel to Dr. Engelmann's monograph, in the 
" Transactions of the St. Louis Academy." 

"The rootstock of all the Yuccas is, under the name of 
' Amole,' an important article in a Mexican household, being 
everywhere used as a substitute for soap, as it is replete with 
mucilaginous and saponaceous matter, probably a substance 
analogous to the saponine of the Saponaria root. It is curious 
to learn that the negroes of the coast of South Carolina 
repeatedly destroyed Dr. Mellichamp's carefully observed 
clumps of Yuccas, in order to obtain the saponaceous root- 
stock." In Colorado Territory we found that Yucca angusti- 
folia is as generally called " Soap-plant " as is the Chloro- 
galum in California. 

While the nature of the fruit, whether capsular or baccate, 
is a tribal character in LUiacecc generally, Yucca has both 
kinds ; and Dr. Engelmann turns this character, with accom- 
panying differences in seeds, to good account in the systematic 
arrangement of the species. The common " Spanish bayonet," 
Y. aloifolia of the Southern States, and some related Texan 
and Mexican species, represent the pulpy - fruited section ; 
Y. brevifolia, which ranges across the Arizonian border of 
the United States, has a spongy indehiscent pod, probably at 
first more or less drupaceous ; while the Bear-Grass, Y.Jihi- 
mentosa, and its allies bear a dry capsule. It may here be 
recorded that the name Y. canaliculata of Hooker must re- 
place that of Y. Treculiana of Carriere, the latter being a 
name published without characters, in 1858, the former de- 
scribed and figured in 1860, in the " Botanical Magazine." 
The prince of Yuccas must be Yucca baccata, which, in its 



RUSKIN'S PROSERPINA. 109 

variety australis, forms " trees twenty-five to thirty feet high, 
and two or three feet in diameter, with ten or a dozen 
branches," or sometimes reaches to even fifty feet of elevation 
according to the late Dr. Gregg, although the most northern 
form of it is almost stemless. Its pulpy fruits are " savory, 
like dates," are eaten fresh by both whites and Indians, and 
are cured by the latter for winter provisions. They also make 
a stew of the flower - buds and flowers, which Dr. Palmer 
found to be pleasant and nourishing. The seeds are said to 
be actively purgative. The fibers of the leaves are used for 
cordage, the trunks for palings, or are riven into slabs for the 
coverings of huts, and the tender top of the stem is roasted 
and eaten. 

Professor Riley's curious paper upon the mutual relations 
of Yuccas with Pronuba, a Tineideous moth that does the 
work of pollination, will be read with interest. 



RUSKIN'S PROSERPINA. 

Mr. Ruskin, " having been privileged to found the School 
of Art in the University of Oxford," now proposes to found a 
new school of botany. 1 Of course, it will be a vagarious 
school. One crying evil to be remedied is " that there are 
generally from three or four up to two dozen Latin names 
current for every flower," and " the most cm-rent and authori- 
tative names " are " of the devil's own contriving." This is 
not seemly. As Wesley would not allow the devil to have 
the singing of all the good tunes, so neither will Ruskin allow 
him to have the naming of all the sweet flowers. He proposes 
" to substitute boldly . . . other generic names for the plants 
thus faultfully hitherto titled." He " will not even name the 
cases in which they have been made," but will " mask those 
which there was real occasion to alter by sometimes giving 
new names in cases where there was no necessity of such kind." 

1 Proserpina, Studies of Wayside Flowers, etc. By John Ruskin. 
London and New York, 1875. (The Nation, No. 528, August 12, 1875.) 



200 REVIEWS. 

That Is to say, the evil of a redundancy of botanical names is 
to be overcome by making more, some of them avowedly need- 
less ; and innocent names are to surfer, lest bad ones should 
become notorious by being discarded without company. For 
it appears that the diabolical names to be sent to their own 
place in this reform are not discarded because they are 
cacophonous, although that is the common charge, but because 
they are immoral. Of the two evils to be dealt with, the first 
is simply a superfluity of Latin names ; the second, and the 
worse, a superfluity of naughtiness. As Mr. Ruskin forcibly 
puts it : — 

" The second, and a much more serious one, is of the devil's 
own contriving (and, remember, I am always quite serious 
when I speak of the devil), namely, — that the most current 
and authoritative names are apt to be founded on some un- 
clean or debasing association, so that to interpret them is to 
defde the reader's mind." 

This reminds us of the fine lady who thanked Dr. Johnson 
for omitting indelicate words from his dictionary, to whom the 
blunt moralist rejoined : " I perceive, madam, that you have 
been looking for them." Now, if the case be really as it is 
represented, the sound practical axiom, " Quieta non movere," 
would seem to suggest the proper treatment ; for his purposes, 
one would think the Latin and Greek names might be left 
untranslated ; and the reform might begin, and end, with the 
popular English names, — almost the only ones the author 
may need to use, — some of which are coarse and vulgar 
enough. Indeed, as to botanical generic names, far from find- 
ing any confirmation of Mr. Ru skin's sweeping charges, we 
can recall barely one or two the translation or etymology of 
which would be embarrassing at the parlor-table. Moreover, 
if the following passages really refer to " nomenclature " 
(though terminology seems to be meant), our author, upon his 
own showing, need not waste his time in the endeavor to 
reform it : — 

" The mass of useless nomenclature, now mistaken for sci- 
ence, will fall away, as the husk of a poppy falls from the 
bursting flower. . . . When the science becomes approxi- 



RUSKIN'S PROSERPINA. 201 

mately perfect, all known plants will be properly figured, so 
that nobody need describe them, and unknown plants will be 
so rare that nobody will care to learn a new and difficult 
language in order to give an account of what in all probability 
he will never see." 

Well, for that matter, the English Botany, in its various 
editions, furnishes fairly good figures of all British plants; 
and the " Botanical Magazine " — a page of an early number 
of which, eighty years old, is gibbeted by Mr. Kuskin — has 
o-one on to figure more than 6000 cultivated exotics, and is 
continuing at the rate of nearly a hundred a year ; so that 
our author's ideal is practically all but realized already. 
There are wellnigh pictures enough, if one knows how and 
where to find them. And it amusingly appears, from Mr. 
Ruskin's trouble with St. Bruno's Lily at the beginning, and 
from his investigation of moss further on, as well as from 
scattered statements, that his mode of proceeding in syste- 
matic botany is the simple one of searching high and low for 
a picture to match the specimen in hand. Accordingly, it is 
not surprising that his " botanical studies were, when [he] 
had attained the age of fifty, no further advanced than the 
reader will find them in the opening chapter of this book." 

As to this, the conclusions which the reader will draw are 
all along anticipated by the author. Next to the pervading — 
well — bumptiousness, nothing is so prominent in the book as 
the profession, not to say the parade, of ignorance of the topics 
treated. As to " the elements of the science of botany," " I 
can scarcely say that I have yet any tenure of it myself." 
" And, meanwhile I don't know very clearly so much as what 
a root is or what a leaf is." " Some one said of me once, very 
shrewdly, "When he wants to work out a subject, he writes a 
book on it. . . . This book will be nothing but processes. I 
don't mean to assert anything positively in it, from the first 
page to the last. Whatever I say is to be understood only as 
a conditional statement — liable to and inviting correction. 
And this the more because, as on the whole I am at war with 
the botanists, I can't ask them to help me, and then call them 
names afterwards." So " for many reasons, I am forced to 



202 REVIEWS. 

print the Imperfect statement, as I can independently shape it." 
To get at the facts, " I should have to write a dozen of letters 
before I could print a line, and the line at last would be only 
like a bit of any other botanical book — trustworthy, it might 
be perhaps, but certainly unreadable." The converse is pre- 
ferred. Keadable it certainly is, and in its way interesting, 
not so much for what it tells about botany as for what it tells 
about Mr. Ruskin ; and the art student, out of the abundance 
of golden chaff, may pick some grains of knowledge that might 
not otherwise fall in his way. But the seeker of botanical 
information must glean warily, especially where the author 
grows positive. For almost the only instance in which he 
does pronounce decidedly happens to be a vexed question in 
vegetable physiology, and there is reason to fear that he de- 
cides it wrongly. At least the recent investigators who have 
had the matter in hand in the way of experimental inquiry, 
will not agree with him that the plant can get water from the 
atmosphere directly and " for the most part does so ; though 
when it cannot get water from the air, it will gladly drink by 
its roots." Still " our natural and honest mistakes will often 
be suggestive of things we could not have discovered but by 
wandering." Very likely ; but why invite learners to go forth 
with him upon his wanderings ? In many a book the want of 
sufficient knowledge is pleaded as an excuse ; in this, it is 
paraded as a recommendation. Ignorance, no doubt, has its 
uses : but it is questionable whether teaching is altogether 
the best use to put it to. As the member of Parliament who 
yawned desperately while delivering his speech was thought 
to trench upon the privilege of his hearers, so the students of 
" Proserpina " may complain that the playing of the role both 
of teacher and learner at once involves some incongruity and 
inconveniences. 

The second part of " Proserpina " has just come to hand. 
It treats of the leaf and the flower, in a discursive and oracular 
way, leading into a3sthetical questions, where we need not 
follow and do not greatly admire. Now and then a scientific 
topic is taken up, and the point missed, as usual. Treating 
of foliage and its office, we are bid " to think awhile of its dark 



RUSKIN'S PROSERPINA. 203 

clear green, and the good of it to you." We look for an ex- 
position of the fact, in which the whole meaning of vegetation 
inheres, that leaves under the sun's influence create all the 
food of the world, and are therefore the basis of all animal 
existence. Instead of which we have : — 

" Scientifically, you know, green in leaves is owing to ' chlo- 
rophyll,' or, in English, to ' green leaf.' It may be very fine 
to know that ; but ray advice to you, on the whole, is to rest 
content with the general fact that leaves are green when they 
do not grow in or near smoky towns, and not by any means 
to rest content with the fact that very soon there will not be 
a green leaf in England, but only greenish-black ones. . . . 
Well, this much the botanists really know and tell us " — 
that vegetation " is made chiefly of the breath of animals. 
... So that you may look upon the grass and forests of the 
earth as a kind of green hoar-frost, frozen upon it from our 
breath, as, on the window panes, the white arborescence of 
ice." 

Mr. Ruskin evidently has no idea of the essential indepen- 
dence of the vegetable kingdom ; that, as all the carbon of the 
breath of animals comes from plants, so they, in their decay, 
would furnish this material for succeeding vegetation perhaps 
as rapidly, on the whole, without the intervention of animals. 
At most, the latter somewhat expedite the decomposition. 

" But how is it made into wood? " As to that and matters 
therewith connected, iC " under the impression that it had been 
ascertained, and that I could at any time know all about it, 
I have put off till to-day the knowing of anything about it at 
all. But I will really endeavor now to ascertain something, 
and take to my botanical books accordingly." 

Behold the result of the cram, "the gist of the matter": — 

"Hence generally, I think we may conclude thus much, 
that at every pore of its surface, under ground and above, the 
plant in the spring absorbs moisture, which instantly disperses 
itself through its whole system ' by means of some permeable 
quality of the membranes of the cellular tissue invisible to our 
eyes even by the most powerful glasses ' ; that in this way 
subjected to the vital power of the tree, it becomes sap, prop- 



204 REVIEWS. 

erly so called, which passes downwards through this cellular 
tissue, slowly and secretly ; and then upwards, through the 
great vessels of the tree, violently, stretching out the supple 
twigs of it as you see a flaccid water-pipe swell and move 
when the cock is turned to fill it. And the tree becomes lit- 
erally a fountain, of which the springing streamlets are clothed 
with new-woven garments of green tissue, and of which the 
silver spray stays in the sky — a spray, now, of leaves." 

Then as to the blossom : " The flower exists for its own 
sake, not for the fruit's sake. . . . But the flower is the end 
of the seed, not the seed of the flower." " The corolla leads 
and is the object of final purpose. The stamens and the 
treasuries [Mr. Ruskin's new term for pistils] are only there 
in order to produce future corollas." Without criticising any- 
body's notion of final causes, we only notice how Mr. Ruskin 
fails to make his own point. He has seen " among the specu- 
lations of modern science, several, lately, not uningenious, and 
highly industrious, on the subject of the relation of color 
in flowers to insects, to selective development," etc. And he 
proceeds to intimate that even Mr. Darwin must be ranked 
among " the men of semi-faculty or semi-education who are 
more or less incapable of so much as seeing, much less think- 
ing about color," etc., referring merely to the ratter's specula- 
tions upon the ocelli of the Argus Pheasant, in blissful igno- 
rance, it would seem, that he has to deal with Mr. Darwin 
upon this very subject of color and use in flowers, and that he 
is not prepared even to state his own side of the question. 



EMERSON'S TREES AND SHRUBS OF MASSACHUSETTS. 

Tree-lore 1 is no longer confined to the few, and books like 
this address a large and various audience, or will do so when 
they become better known. Mr. Emerson's original volume 

1 A Report on the Trees and Shrubs of Massachusetts, growing naturally 
in the forests of Massachusetts. By George B. Emerson. 2d ed. Boston, 
1875. (The Nation, No. 539, October 8, 1875.) 



TREES AND SHRUBS OF MASSACHUSETTS. 205 

was published twenty-five years ago, " agreeably to an order 
of the Legislature, by the Conimissiouers on the Zoological 
•and Botanical Survey of the State " of Massachusetts, being 
a supplement to the Geological Survey of that period, prose- 
cuted under Edward Everett's governorship. It was the most 
popular report of the series, and the edition was ere long 
exhausted. When this came to pass — without waiting for 
the new survey which the State last year came near authoriz- 
ing but failed to do so — Mr. Emerson, unassisted, set about 
the preparation of a new edition, devoted several years to it, 
and to the study of what had been done for the preservation 
and utilization of forests in the Old World, and for their 
waste and destruction here in the New ; and he has at length 
brought out this second edition, in two goodly octavo volumes, 
illustrated and adorned by a large number of well-executed 
plates. These being interspersed through the pages, unnum- 
bered, and nowhere enumerated, the only way of ascertaining 
their actual amount was to count them. We find 144 plates, 
of various kinds and merits. The least satisfying to us 
are those of portraitistic or scenic character, borrowed from 
the German " Der Wald " and the very French " Vegetable 
World " of Figuier ; yet to others these may be the most 
attractive. Very good, though unpretending, are figures, 
mainly in outline, contributed by Mr. Isaac Sprague to the 
first edition, here reproduced. Best of all are those contrib- 
uted by the same hand to the new edition, original figures of 
the foliage, flowers, and fruit of many of our trees and shrubs 
not before illustrated, transferred from Sprague ? s drawings to 
stone, and printed in colors. The plates representing our two 
northern Azaleas, the Roxbury Waxw r ork (as they name it 
around Boston), the Virginian Creeper in its autumn dress ; 
the Red Maple, both in vernal and autumnal robes ; and the 
Flowering Raspberry, from which seemingly one may almost 
shake the mountain dew, are good illustrations of what may 
be done in this way. Hand-coloring is too expensive, and 
chromo-lithography can really be turned to excellent account 
in its place for natural-history illustrations, whatever be its 
merits or demerits in other regions of art. The letter-press 



20G REVIEWS. 

we find Is a reprint of the first edition, as to the descriptive 
part. The popular descriptions seem to have hit the mark. 

Mr. Emerson's instructions and appeals for the planting 
and care of trees, and for the renovation of our woods, wher- 
ever practicable and profitable, are worthy of all attention, as 
his efforts in this regard are worthy of all honor. They began 
long ago, and have been redoubled now in his later years, in 
this work and elsewhere. The memorial addressed to the 
President and Congress by the American Association for the 
Advancement of Science he took a large part in preparing ; 
and his personal furtherance of it at Washington, of which 
he makes modest mention in the preface, may yet be fruit- 
ful of benefit. It may not be improper to add that the only 
permanently endowed arboretum in America — the Arnold 
Arboretum, entrusted to Harvard University — owes its ex- 
istence to our author's thoughtfulness and sense of the im- 
portance of tree-culture. 



DARWIN'S INSECTIVOROUS PLANTS. 1 

This long expected work appeared last autumn, was imme- 
diately reprinted by the American publishers, and before this 
time has been so widely read that no detailed account of it is 
at all necessary. Its main topic is Drosera or Sundew, upon 
which the vast number and diversitv of the observations and 
experiments — at once simple, sagacious, and telling — which 
it records, are about as wonderful as the results. As to the 
latter, it is established beyond question that the common 
Sundews are efficient fly-catchers ; that the stalked glands, or 
tentacles as Mr. Darwin terms them, are sensitive and turn 
inward or even in other required directions in response to 
irritation ; that they equally respond and move in obedience 
to a stimulus propagated from a distance through other ten- 
tacles and across the whole width of the leaf ; that the sensi- 

1 Insectivorous Plants. By Charles L»arwin. London and New York, 
1875. (American Journal of Science and Arts, 3 ser., xi. 69.) 



DARWIN'S INSECTIVOROUS PLANTS. 207 

tiveness belongs only to the glands and tips of the tentacles, 
but is propagated thence down their stalks and across the 
blade of the leaf through the cellular tissues ; that they accu- 
rately and delicately discriminate animal or other nitrogenous 
matter from anything else ; that the glands absorb such mat- 
ter ; that when excited by contact, or by the absorption of 
nitrogenous matter by the viscid enveloping liquid, an acid 
secretion is poured out and a ferment analogous to pepsin, 
the two together dissolving animal matter ; so that the office 
and action of these glands are truly analogous to those of the 
glands of the stomach of animals. Finally, that animal or 
nitrogenous matter, thus absorbed and digested in the glands, 
is taken in, and conveyed from cell to cell through the tenta- 
cles into the body of the leaf, was made evident by ocular 
inspection of the singular changes in the protoplasm they con- 
tain. So particularly have the investigations been made and 
so conscientiously recorded, that the account of those relating 
to one species of Sundew, Drosera rotund [folia, fills 277 
pages of the English edition, or more than half of the book. 
After all it ends with the remark : " and we see how little 
has been made out in comparison with what remains unex- 
plained and unknown." The briefer examination of six other 
Sundews follows, some of them equally and others less effi- 
ciently fly-catchers and feeders. 

Dionaea is next treated, but with less detail. Indeed, ex- 
cept as to the particular nature of the secreted digesting fluid, 
there is little in this chapter that had not been made out or 
already become familiar here. That the secretion has diges- 
tive powers, and that it is reabsorbed along with whatever 
has been digested, is now proved beyond reasonable doubt. 
That the motor impulse is conveyed through the cellular par- 
enchyma, and not through the vascular bundles, or spiral ves- 
sels, and that the latter do not originate the secretion, as Rees 
and Wills in a recent paper seem to suppose they must, ap- 
pears to be shown by the facts, and was antecedently probable. 
" The wonderful discovery made by Dr. Burdon Sanderson is 
now universally known : namely, that there exists a normal 
electrical current in the blade and footstalk, and that when 



208 REVIEWS. 

the leaves are irritated the current is disturbed in the same 
manner as takes place during- the contraction of the muscle of 
an animal." The conclusion here needs to be checked by 
parallel experiments, to see whether the same reversion of 
current does not take place whenever a part of any leaf or 
green shoot is forcibly bent upon itself. 

Aid formula vesiculosa, of the Drosera family, "maybe 
called a miniature aquatic Dionaea ; " for, as discovered by 
Stein in 1873, " the bilobed leaves open under a sufficiently 
high temperature, and when touched suddenly close." Being 
submerged, their prey is confined to minute aquatic animals. 
For want of proper material and opportunity, Mr. Darwin 
was able to follow up only for a little way the observations of 
Stein and Cohn, — enough, however, to show that it also cap- 
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 Drosera Jil {for mis, 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 ordinary glandular hairs of several plants. 
To certain Saxifrages his attention was naturally called, on 
account of the presumed relationship of Droseracece 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 asan 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 Diona3a and Drosera — the bladders 
imprisoning minute aquatic animals by a mechanism almost 
as ingenious as that of Dionaea itself. Observations of the 
same kind were made in this country by Mrs. Treat, of Vine- 
land, New Jersey, before Mr. Darwin's investigations w r ere 
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, however, according to Dr. Hooker's investi- 
gations, has attained a higher dignity, and converted its pitcher 
into a stomach. This parallelism, and this higher aud 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 which 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 

the pitcher of Sarraceniajlava, 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 
Linnaean Society. His observations (made upon S. vario- 
lar is^), it appears, were communicated to Sir J. E. Smith, 
read before the Linnaean Society in 1815, and published soon 
after. They are referred to by his surviving friend and 
associate, Eliott, in his well-known w r ork, 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 faux 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 REVIEWS. 

Dr. Macbride'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 
VARIABILITY IN PLANTS. 1 

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. 

1 American Journal of Science and Arts, 3 ser., xi. 153. 



HEREDITY AND VARIABILITY IN PLANTS. 213 

The " Comptes Eendus " of September 27 and October 
4, 1875, contain an abstract of a paper communicated by M. 
Naudin to the French Academie des Sciences, of which the 
text was suggested by a hybrid between the wild Lactuca 
virosa and a variety of L. 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 parental heritage, no two being quite 
alike. This exceeding vacillation between the two parental 
forms, but not overpassing the limits on either hand, — which 
Xaudin finds to be the common characteristic of fertile 
hybrids, close-bred, — he names disordered variation (variation 
desordonnee). His explanation is that the hybrid is a piece 
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 place, part of the 
progeny reverting completely to one ancestral type, part to the 
other. "What Xaudin 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 
consecpience 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 oris-- 
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 per se, by making a cross, 



21-4 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 desordoti?iee, which would 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 which 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 enndo ; 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 apprehend 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 appear 
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 resumption 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-organism " ; and that 
the various lines of descent acquired fixity into species only 
as they acquired sexuality. If we rightly 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 plants propagated from buds. This we 
suppose can be answered only one way. 

M. Naudin is a veteran and excellent investigator, and 
nothing which he writes is to be slighted. We have frankly 
set down our impressions upon a first perusal of bis 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 bis new work, 1 refers only 
incidentally to adaptations for cross-fertilization, — a subject 
wbicb has given origin to a copious literature since he opened 
it anew in bis 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 and Self -Fertilization in the Vegetable Kingdom. 
By Charles Darwin. London and New York, 1876. (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. 

Ipomo&a purpurea, 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 IN 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 tbe 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 opposite 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 jDair 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 an 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 average 
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, anil 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 
generations taken together is less than in the first and second 
generations together." Upon this Mr. Darwin remarks : 
ik When, however, we remember that the self-fertilized and 
crossed plants 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 Ipomrea, while 
other flowers 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 
flow r ers 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 Ipomsea 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 



22-2 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 
Eschscholtzia, 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 purpurea, 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 plants 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 



FERTILIZA TION 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 Mimulus 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, though small advantage appeared to come from 
the crossing of flowers on the same plant. 

In Origanum vulgare, crosses were 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 
growth those of the self-fertilized ; " a cross of this kind did 
no more good than crossing two flowers on the same plant of 
Ipomsea 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 produced 
thirty flower-stems, while the self-fertilized produced only 
fifteen, or half the number. The pots were then bedded out, 



224 REVIEWS. 

and thf 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, — somewhat 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 
Ipomasa, 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 self-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 Exchscholt\i<i 
( '<il[fornica 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 of 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 
crown 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 genei'ations, 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 chapters 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 periods; 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 prepotency of pollen from 
any other individual over that from the same plant. AVe 
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 OrchideoB in both being excluded. If the genera 
in this latter order, as well as in the Asclepiadeaz and Apocy- 
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, however, 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 Trifolium repens and T. pratense 
(the common "White and Eed Clovers) have a place in the first 
list; T. arvense and T. procumbens 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 Hymenoptera, 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 Impatiens " 
(for which Gould, " Trochilidse," 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 ( Tecoma 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 
leaving 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 depend 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 flowers : 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, 
which 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 weather. 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 pollen, in comparison 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,G00, 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- 
?nophilous, 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 
cleistogenons 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, Viola, and the like, it is provided for by separate 
flowers, the special adaptations of which are unmistakable. 

H. Muller 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 
quickly 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-fertilizable. 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 
pollen." 

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- 
in^. So does the remarkable number of flowers which bees 
are able to visit in a short time (of which mention will 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 
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 twenty 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. 

\\e 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 though growing 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, would 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 proportion 
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 phaenogainic vegetation, 
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 JIarcgraviacece. 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, which 
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 Conifercc, Birches, Beeches, etc., or meadows and glades, 
as of Grasses, Sedges, and Rushes. 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 Algce : 
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. Midler 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 shall presently see that this holds good in their felo- 
nious practice of biting holes through the corolla." — (p. 420.) 
As to this hitter 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 play 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), w r hilst 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 eniploj'ed in cutting holes in flower after 
flower ; and on the next day every single hive-bee, without 
exception, instead of alighting on the left w T ing 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 fur many following days. Mr. 
Belt has communicated to me (July 28, 1874) a similar ease, 



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 quickly 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 high 
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 specialization and 
perfection. 



238 REVIEWS. 



THE HYBRIDIZATION OF LILIES. 

Mr. Paukmax, under the above title, gives a summary 1 of 
his experiments, 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 L. 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 IAlium Parlc- 
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 parent 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 was enough to produce 8 or 10 young bulbs. Of 
these, w T hen they bloomed, one bore a flower combining the 
features of both parents, but though large, it was far inferior 
to L. Parkmanni 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 attenrpts to 
hybridize other Lilies. For instance, our L. superbum was 
pollenized 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. superbum." 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 nfaternal 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, L. superbum 
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. speciosum, 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 wei*e 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 L. superbum 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 the 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. 241 



PHYTOGAMY. 

If this name has not been coined already it ought to be. 
For " the loves of the plants," so niellifluously 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. 1 
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 bim, 
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 
brought to it must come from the pin-eyed, and vice versa, in 
order to secure full fertility. Tiny blue-eyed Houstonias, en- 
amelling our meadows in early spring, and fragrant Mitchel- 
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 by which Orchids are Fertilized by Insects. Second 
Edition, revised. — The Effects of Cross and Self-Fertilization in the Vege- 
table Kingdom. By Charles Darwin. London and New York, 1876-77. 
(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 ilowers — 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 flower 
it belonged to, but potent upon the stigma of the other sort, 
upon 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 dimorphism 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 cori-elations 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 pistils 
occupy distinct plants or different blossoms. To these three 
classes, then, Mr. Charles Darwin's aphorism, " 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 surprising 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 
anticipated 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 which 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. Readers 
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. "While 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, 
cceteris ])aribus, hi 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 flow r ers 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 principle 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, we 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- 
togamous 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 1 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 plnenogamous Flora of a continent, and the only one ; 

1 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 
srood botanical work at such a rate. If, as some fear, the 
race of first-class systematic (phsenogamous) 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, Von 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 Tasmania?,' ! 
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 Graminece, 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 Mr. 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 Cyjwracece, 
upon the arrangement of which sound judgment is brought 
to bear. The great order Liliacece is made to include the 
Smilacece, and not the Roxbtirghiacece. "We should have 
excluded both, but Smilax in jjreference. Contrary to Mr. 
Bentham's opinion, we should insist that the anthers in 
Smilax are unilecular but bilocellate. The diagnosis of It ox- 
burghiacece in the conspectus distinguishes the order from 
Australian Liliaceaj 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 Hestiacece, by Dr. Masters. This 
is an order allied on the one hand to Juncacece, on the other 
to Cyperacece, of twenty genera and two hundred and thirty- 
four species, wholly of the southern hemisphere, divided 
between South Africa (which has much the larger share), 

1 Monographic^ Phanerogamarum Prodromi nunc continuatio, nunc revi- 
ew, auctoribus Alphonso et Casimir De Candolle, aliisque Botanicis ultra 
memoratis. Vol. I. Smilacece, Restiacece, Meliacece, cum tabulis ix. Paris, 
Jane, 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 systeinatist, 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 Jfeliacece, 
chiefly a tropical order. The stamineal tube in the monadel- 
phous Jleliacece 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 
monadelphous stamens (east Asiatic), Smilax with separate 
sepals, petals, and (6-15) stamens ; the third, Rhipogonum 
(of Xew Zealand and Australia), with he rm aphrodite 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 Candolle's 
best manner. "We are pleased to find that he keeps up the 
sjjecijir phrase, and with true Linnsean 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 stipides, 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 raisoji 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, ami with dull-colored perianth, the pollen is 
papillose as in most entomophilous flowers ; but that lihipo- 
gonum, 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 whether 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. c, all the types of the order — coexist in the 
comparatively small area comprised between the north of New 
Holland, the Fiji Islands, the Sandwich Islands, and Japan ; 
that India has four of these six types, New Holland three, 
North America two, all Europe and Africa one ; South Amer- 
ica only one, but is rich in species. The speculative inference 
is, that, anterior to the eocene formations of Europe, the 
ancestors of the family occupied a continent situated, in the 
region above indicated, of which the most ancient form was 
probably monoecious, gamosepalous, apetalous, monadelphous, 
and with more or less volatile pollen, — in short was like He- 
terosmilax ; that this ancestor was in that region diversified, 
giving origin to the five other groups, beginning with Eu- 
smilax, the widest diffused and most numerous in species, 
and finishing with Rhipogonum, which with Heterosmilax 
has clung to its birthplace. The sole Californian Smilax is 
referred, as a variety, to S. rotund i folia, but is nearer S. 
liispida, although distinct from both. 

Vol. V. Pars secunda : Ampelidece ; by J. E. Planchon, has 
at length appeared. It occupies 350 pages ; and it represents 
a great amount of labor, the permanent value and complete 
acceptance of which cannot be adjudicated off-hand. The 
plan of merging all the forms into one genus, Vitis, has been 
abundantly tried, not with very satisfactory results, — partly, 
it may be, because the groups have not been well worked 
out. Professor Planchon, a most experienced and keen bot- 
anist, who has especially investigated the Vines for a good 



DE CANDOLLE'S NEW MONOGRAPHS. 251 

many years, has very naturally tried the other tack, and has 
developed the Linnaean Vitis and Cissus into ten genera. The 
principles upon which he has proceeded, as explained in the 
preface, are wholly legitimate ; and one could wish that they 
have been successfully applied. This, only use can deter- 
mine. We may be confident, however, that if this monograph 
had been in the hands of the authors of the latest Genera 
Plantarum, they would not have bodily adopted its conclu- 
sions, although they would have been much helped by the 
elaborate investigations, and might have seen their way to 
admit three or four genera. They would not have trusted 
over-much to the difference between polygamo-dicecious, poly- 
gamo-monoecious and partly pseudo-hermaphrodite, herma- 
phrodite and probably some pseudo-hermaphrodite, and her- 
maphrodite or rather physiologically polygamo-monoecious and 
with some blossoms pseudo-hermaphrodite, — differences which 
must be shadow}-, — nor to variations in the mere shape of 
style and stigma. And as to the disk, which should be more 
tangible and hopeful, we gather from Planchon's synopsis and 
from our own observations that there are only three types. 
In the true Grapevines the disk is represented by nearly dis- 
tinct and free nectariferous glands, alternate with the stamens. 
In most other AmpelideoR, it is cupular or annular (entire or 
crenate or lobed), with base or lower half more adnate to the 
base of the ovary, but at least the margin or lobes free. In 
the Virginia Creeper there is really no disk at all, as was first 
noted by Dr. Torrey in his " Flora of the Northern States," 
in 1824, and insisted on in the " Flora of North America," in 
1838, and again in the " Genera Illustrata," where there are 
correct figures. Dr. Planchon expresses the same opinion in 
essence but in different language, i. e., "Discus obsoletus 
ovarii basi plane adnatus et tantum colore proprio subdis- 
tinctus." We could not make much of the color ; but the 
tissue does thicken more or less, and possibly may become 
obscurely nectariferous ; but the flowers are not attractive to 
bees, as the allied species from Japan is. In the latter, while 
there is equally no hypogynous disk, there is much thickening 
of nectariferous tissue over all the lower part of the ovary 



252 REVIEWS. 

more or less in longitudinal ridges, the whole " plane adnatus " 
throughout. Now we should make more of these three types 
than Dr. Planehon does. For the first goes with the calyp- 
trately caducous corolla and polygarno-dioecious flowers of 
true Vitis. The third with disk, if so called, wholly confluent 
with the ovary itself, belongs to and includes all of the few 
known species (including Planchon's Landukia), which have 
the striking biological character of climbing by the dilatation 
and adhesion of the tendril tips ; and their flowers are 5-mer- 
ous, essentially hermaphrodite, and with expanding corolla. 
The second type of disk goes with 4-merous and some 5-merous 
flowers with corolla expanding in anthesis, that is, to the genus 
Cissus. We do not see the way to break this up into genera, 
certainly not on the number of parts, for this varies in some 
species, and while C. stems is 5-merous, the closely related 
C. orientalis is 4-merous. However it may be with some 
exotic groups, we must restore our two species, which formed 
part of Michaux's Ampelopsis, to the genus Cissus. Under 
that view the generic nomenclature is clear. The genus Am- 
pelopsis (Michaux, p.p. and Torr. and Gray) is to be main- 
tained on the lines long ago laid down in this country, and 
now reinforced, for those species which are popularly well 
known under this name. AYe do not feel obliged to defer to 
any work of Rafinesque as late as the year 1830. But, as to 
the present point, it seems to us that when Dr. Planehon fol- 
lowed him in the appropriation of one part of Michaux's Am- 
pelopsis, he should also have adopted Rafinesque' s name for 
the other part, namely, Quinaria, instead of making a new 
name, Parthenocissus, the former name being free for use. 
In our view both names are superfluous. As to true Vitis, 
it remains to be seen whether it will be at all possible to dis- 
tinguish twenty or more North American species. Perhaps 
Engelmann allowed quite as many as can be defined. But 
Planchon's long and conscientious labors upon the genus and 
the family must be most helpful even where his conclusions 
are not at once accepted. 1 

1 This is the last Review written by Professor Gray. — C. S. S. 



EPP1NG FOREST. 253 



EPPING FOREST. 1 

" Eppixg Forest, and How best to deal with it," is an article 
by Mr. Wallace in the "Fortnightly Review" for November, 
also separately issued, which should not be passed over as a 
matter of local concern. All Mr. Wallace's writings, even 
the most casual, will be found to touch and to illustrate some 
interesting question. A recent act of Parliament having de- 
creed that one of the ancient woodland wastes, Epping For- 
est, which lies upon the very borders of Loudon, shall be pre- 
served forever as " an open space for the recreation and 
enjoyment of the public," the question what to do with it be- 
comes a very practical one. It is not enough to say rejoic- 
ingly : " Here at length every one will have a right to roam 
unmolested, and to enjoy the beauties which nature so lavishly 
spreads around when left to her own wild luxuriance. Here 
we shall possess, close to our capital, one real forest, whose 
wildness and sylvan character is to be studiously maintained, 
and which will possess an ever-increasing interest as a sample 
of those broad tracts of woodland which once covered so 
much of our country, and which play so conspicuous a part 
in our early history and national folk-lore." Unfortunately 
much of it has been spoiled in all senses of the word. But 
Englishmen know how to plant, and the native trees which 
once covered the domain, with the undergrowth which of old 
accompanied them, could be made to flourish again. Prob- 
ably the ancient forest could be essentially reproduced in all 
its former vigor, and former monotony. Mr. Wallace has 
something better than this in his mind, and his inspiration is 
caught from Professor Asa Gray's Harvard lecture on " For- 
est Geography and Archaeology," which was published last 
summer in the " American Journal of Science," the ideas of 
which he adopts, happily summarizes, and applies to the case 
in hand. In re-foresting the open waste portions of Epping, 
he proposes to establish several distinct portions or broad 
tracts, each composed solely of trees and shrubs belonging to 
1 The Nation, No. 701, December 26, 1878. 



254 REVIEWS. 

some one of the great forest regions of the temperate zone. 
A climate of which it has, we believe truly, been said that it 
can grow treble the number of species of trees which the At- 
lantic United States can, and in which so many trees have 
been individually tested, offers favorable auspices for an 
undertaking of this kind upon a scale that may give a good 
idea of the features — not of this or that tree or shrub, but 
of a forest of the Alleghanies, of the Sierra Nevada, of Brit- 
ish Columbia, and of Japan. Even the southern temperate 
zone may contribute from New Zealand its Kauri Pines and 
Beeches, under which Macaulay's overworked New Zealauder 
may encamp on returning from his excursion to view the ruins 
of London bridge by moonlight. 

"When Mr. Wallace declares that " there is really no diffi- 
culty in producing in England an almost exact copy of a 
North American forest, with all its variety of foliage, with 
its succession of ornamental flowers, and with its glorious 
autumnal tints," we must agree that the experiment as a 
whole is hopeful, and much of it is already a success in piece- 
meal plantation. But we are not sure about autumnal tints 
under London skies, considering how much these differ be- 
tween one season and another in New England. And, though 
every tree will grow in England, being put to no severe stress 
either in winter or summer, yet not every tree nurtured under 
our climate — so fierce in both seasons — will blossom in 
England, as witness our handsome leguminous tree, Cladrastis, 
or Yellow-wood. But a climate which will fairly nourish on 
one soil the trees of the Atlantic and the Pacific forests, 
those of Japan and Mantchuria, of Siberia, Himalaya, and 
the Caucasus, along with those to the manner born, deserves 
to possess them all. We, alas ! can seldom grow on one side 
of our continent the trees and shrubs of the other. More- 
over, there is very little forest east of the Rocky Mountains 
which an act of Congress could preserve ; and, over that lit- 
tle, Congress and the Secretary of the Interior have lately 
been at loggerheads. Yet in California we have forests, still 
public domain, which are the veritable wonders of the world, 
which for the most part are doomed to irremediable destruc- 



HOOKER AND BALL'S TOUR IN MAROCCO. 255 

tion, but of which specimens ought to be preserved now when 
they may, now when it will cost nothing, and injuriously af- 
fect no man's interest. For the Redwood it is almost too 
late ; yet a square mile, or half that area, of Redwood forest 
might still be reserved in Mendocino or Humboldt County. 
And from all accounts a square mile or two of true Big-tree 
forest, on or south of King's River, could well be set apart 
as a perpetual memorial. The Mariposa Grove is indeed 
such a reservation. But this is only a grove of a limited 
number of trees, many of them sadly injured by fires. Far- 
ther south this great tree is said to be the main constituent of 
extensive forests. A mile or two of Big-tree forest should be 
set apart before this district is invaded. 



HOOKER AND BALLS TOUR IN MAROCCO. 

The isolation of Marocco from the European world is strik- 
ingly shown in the fact that up to the publication of this 
goodly volume 1 the principal geographical authority for the 
interior of the country is Leo Africanus. This man was a 
Moor of Granada, who with his kinsfolk, at the time of the 
siege of Granada in 1492, fled to Fez, then the headquarters 
of Arabic culture, and was taken into favor by Mouley Ahmet, 
the founder of the dynasty still reigning in Marocco. He 
traveled throughout the empire, and wrote in Arabic a descrip- 
tion of that part of Africa, which he must have had with him 
in manuscript when, in the year 1517, he was captured by 
Christian corsairs, and carried to Rome. The Pope, Leo X., 
hearing of this learned Moor, sent for him, treated him kindly, 
had him baptized, and gave him at the font his own names, 
Giovanni Leone ; and so, as the writer of the narrative before 
us naively remarks, the Moor perhaps became as earnest a 
Christian as the pontiff himself. In Rome he translated his 

1 Journal of a Tour in Marocco and the Great Atlas. By Sir J. D. 
Hooker and John Ball. London and New York, 1878. (The Nation, 
No. 718, April 3, 1879.) 



256 REVIEWS. 

work into Italian ; bnt it was not printed until 1550, in 
the first edition of Ramusio's " Collection of Voyages and 
Travels." Excepting Gerhard lloklfs — who, by assuming 
the garb and professing the faith of a Mussulman, traversed 
districts where no Christian dare present himself, and who 
could make only surreptitious observations and trust them to 
memory — our British travelers were the first Europeans to 
travel in South Marocco and to reach the Great Atlas. 

Their opportunity and their success were exceptional. It 
was a botanical expedition, in the spring of 1871, with the 
president of the Royal Society and director of Kew Gardens 
at its head, with Mr. Ball, a member, or at least an ex-member 
of Parliament, as his companion, and a third naturalist, Mr. 
George Maw, as their associate for half the journey. High 
influence was required for obtaining the Sultan's permission to 
penetrate the country at all ; and utmost skill, determination, 
and no small assertion were called for to surmount the obstacles 
which were systematically interposed by the authorities, and 
to baffle attempts to lower the personal importance, and there- 
fore frustrate the aims, of this scientific embassy. On reach- 
ing the city of Marocco the playing of the game began. The 
Sultan's letter instructed the governor of Mogador to " send 
the English hakim and his companions to the care of my slave 
El Graoui," said slave being the governor of the whole Great 
Atlas region. The city of Marocco was not in his province ; 
and there the duty of providing for the sustenance and com- 
fort of the travelers devolved upon Ben Daoud, governor of 
the city. Before entering the town our travelers had managed 
to learn that a very small house with only two rooms had been 
provided for them, and that Ben Daoud meant to make as 
little of them as jjossible. So a message was sent on that a 
larger house was needed, or else an enclosed garden in which 
to pitch their tents. In reply, a larger house with four rooms 
was offered. On reaching this house at nightfall it proved to 
be mean, dirty, and swarming; so when the mo?ia, or present 
for the evening meal, came in, it was ordered back with scorn. 
It was felt that submission to any belittling at the outset would 
be the beginning of sorrows. 



HOOKER AND BALL'S TOUR IN MAROCCO. 257 

" Tell the Governor," said Hooker, " that my Sultana gives 
me a large house with a garden to live in ; hospitality would 
require that the Governor of Marocco should provide me — 
the guest of his Sultan — with a better house ; but, in any case, 
I shall not live in a worse one." 

The messenger returned with the answer : " The Governor 
has no better house to give the Christians ; but Marocco is 
large, and they are welcome to provide for themselves." 
Whereupon the cavalcade moved to the great square, or open 
space, beside the chief mosque and tower of the Kontoubia, 
sending at the same time a message to the Viceroy, son of the 
Sultan, that they should encamp in their own tents until a 
suitable house had been provided. The upshot was that on 
the following day the Viceroy installed them in the palace of 
Ben Dreis, with the adjoining garden, and soundly berated 
the city governor for his churlishness ; but it leaked out that 
the poor governor had ouly obeyed the express orders of the 
Viceroy, who had directed him to begin by offering the mean 
house, then one somewhat larger, and to leave it to the Vice- 
roy himself graciously to meet the higher demands if they 
should be insisted on. Fortunately, too, the row with the city 
governor threw the travelers into the hands and good graces 
of his rival, El Graoui, under whose protection and care they 
were to explore the Atlas range. 

It was now full time to determine how this visit and the 
laborious journey they were to undertake were to be made 
intelligent and satisfactory to El Graoui and the other Moor- 
ish authorities. The gratification of a desire to learn some- 
thins: about the vegetation of the Great Atlas would have 
seemed a thin pretext for some sinister design. An insanity 
of this sort might possess the unaccountable soul of some one 
Christian ; but that three should be simultaneously smitten 
with it would be thought to pass the bounds of probability. 
The pretense of collecting live plants for Kew Gardens could 
hardly be made plausible to the Moorish mind, except, haply, 
for that one use of plants that every one can understand. The 
Royal Gardens are literally the Queen's Gardens, and herbs 
that will cure diseases are among its most valuable collections. 



o 



258 REVIEWS. 

Judicious representations upon this line would go far toward 
explaining the strange proceeding of the party of travelers, 
and would contain all the truth they were likely to get credit 
for. So ki there is no doubt that the current belief among our 
own followers was, that the Sultana of England had heard that 
there was somewhere in Marocco a plant that would make her 
live forever, and that she had sent her own hakim to find it 
for her." And when it was seen what toil and hardship these 
botanical explorations entailed, the natural commentary upon 
the whole proceeding was : " The Sultana of England is a 
severe woman, and she has threatened to give them stick 
(the bastinado) if they do not find the herb she wants." 

Those interested in the natural history of an almost un- 
known district lying on the very border of Europe ; those 
interested in its geography and physical features ; the states- 
man and philanthropist, who may here contemplate in a typi- 
cal instance the decadence and depletion of a once prosperous 
people, and note the misery which tw r o or three centuries of 
misrule may bring upon a smiling land ; and those who enjoy 
the charm of fresh and graphic narrative, the sketch by a 
practised hand of keen and quick observations by practised 
eyes, — all these will find this volume attractive reading. To 
us it has been fascinating and full of suggestion. The main 
drawback to our travelers' enjoyment was the suffering which 
they unwittingly inflicted on the mountain villagers of the 
Atlas — the poor Berbers, who are altogether the worthier 
part of the population — through the insatiable rapacity of a 
large escort provided by the Moorish government for their 
protection. If the naturalists could have journeyed by them- 
selves, with needful attendants under their complete control, 
their support would have entailed no hardship whatever ; but 
the shameless extortion and abuse by the guards, which the 
travelers were unable to prevent, must have left a painful 
remembrance upon their minds as well as upon those of the 
despoiled mountaineers. It is hardly a consolation to the 
former that the latter are used to it. 

The appendix to this charming volume is filled with scien- 
tific, philological, and geographical details. The most elabo- 



BENTHAM ON EUPHORBIACEJE. 259 

rate article is that " On the Geology of the Plain of Marocco 
and of the Great Atlas," by Mr. Maw. 

As to the orthography of the name of the city and country, 
Morocco is peculiar to the English. In the adoption of 
Marocco the authors follow, as regards the first vowel, the 
universal continental usage. 



BENTHAM ON EUPHORBIACE^E. 

This thoughtful essay 1 presents the general views attained 
to by Mr. Bentham on working up the genera of the great 
order Evjihorbiacece for the ensuing volume of the " Genera 
Plantarum." AVe need not specify any of the results, except 
to indicate the author's decision in the case of the Buxece. 
He does not follow his predecessors, Baillon and J. Mueller, 
who, much as they differ in other respects, agreed in setting 
up the order Bvxjiccrv taking their cue from Agardh, and 
making much of the dorsal rhaphe. Bentham concludes that 
this small group, however well defined, ought not in a general 
view to be regarded as of higher grade than one of the pri- 
mary divisions, or tribes, of Euphorbiacece. AVe are not the 
less pleased with this that we quite expected it. 

A wider interest will be felt in Mr. Bentham's cxctirsus on 
nomenclature, or rather on some questions which the study of 
Evpliorbiacece brought up, and which some recent discussions 
have made pertinent. The general laws of nomenclature of 
our day, and the principles on which they rest, are laid down 
in the code which was reported by Alphonse De Candolle 
to the Paris International Convention, in the year 1867, 
and, being approved, was published with a commentary in 
the autumn of that year, and in an English translation early 
in the following year. The laws, without the commentary, 
were printed in this Journal for July, 1868. The ten years 

1 Notes on Euphorliacece. By George Bentharu ; Journal Linnsean 
Society, xvii. London, 1880. (American Journal of Science and Arts, 
3 ser., xvii. 335.) 



260 REVIEWS. 

succeeding have tested, somewhat thoroughly, the questions 
(nearly all of minor moment) upon which differing usages 
prevailed ; and though one or two points are still mooted, the 
great majority of plnenogamous botanists are coming to be 
of one mind and practice. But, as Mr. Bentham remarks : 
" The result has not been quite effectual in checking the ever- 
increasing spread of confusion in synonymy. Besides the 
young liberal-minded botanists who scorn to submit to any 
rule but their own, there are others who differ materially in 
their interpretation of some of the laws, or who do not per- 
ceive that in following too strictly their letter instead of their 
spirit, they are only adding needlessly to the general dis- 
order. In the application as well as in the interpretation of 
these rules they do not sufficiently bear in mind two general 
principles : first, that the object of the Linnaean nomencla- 
ture is the ready identification of species, genera, or other 
groups for study or reference, not the glorification of bota- 
nists ; and secondly, that changing an established name is 
very different from giving a new name to a new plant." 

It is to the latter point that this most experienced and 
even-minded botanist addresses himself. " The rule that long- 
established custom amounts to prescription, and may justify 
the maintenance of names which form exceptions to those 
laws which should be strictly adhered to in naming new 
plants, is unfortunately now frequently ignored. . . . The 
law of priority is an excellent one ; and when a genus or 
species has been well defined by an early botanist in a gen- 
erally accessible work, but has subsequently been neglected, 
and the plant became known under other names, it is well 
that the original one should be restored. . . . On the other 
hand, it creates nothing but confusion to suppress a generic 
name, well-characterized and universally adopted by long 
custom, in favor of a long-forgotten one, vaguely designated in 
an obscure work, out of the reach of the great majority of 
botanists. . . . The greater number of Necker's genera have 
been so imperfectly characterized, with so absurd a terminol- 
ogy, that they are quite indeterminable ; and his names de- 
serve to be absolutely ignored, except in the very few cases 



BENTHAM ON EUPHORBIACEJE. 261 

where Jussieu or other early French, botanists have succeeded 
in identifying- them, and corrected their characters ; but even 
then it is doubtful whether these names should not bear the 
date of the correction, rather than of the original work. 
Adanson's ' Families,' with all the inconveniences of its 
form and absurd orthography, is much more scientific, and 
many of his genera are well defined, and have therefore been 
properly adopted." . . . 

Let us here interject a practical application. There is an 
old and well-established genus Smilacina of Desfontaines. 
There is a much older genus Tovaria of Ruiz and Pavon, 
founded in 1794, ever since accepted, and without a syn- 
onym. Recently Mr. Baker of Kew, finding that Necker has 
a Tovaria, published in 1790, and therefore four years earlier 
than that of Ruiz and Pavon, takes up this name in place of 
Smilacina, and leaves a new name to be made for the long- 
established homonymous genus. It will be said that the rule 
of priority demands the sacrifice, and that the identification 
of Necker's genus is sure, because the three Linnsean species 
of Convallaria which properly constitute Desfontaines' Smi- 
lacina are referred to it by name ; and that, though it be a 
case of snmmum jus summa injuria, the injurious conse- 
quence is a necessity. But Mr. Bentham's characterization 
of Necker's work applies even to this instance. Twice over 
Necker's Tovaria is described as having a perianth of five 
sepals, and the berry is said to be one-celled. Desfontaines' 
Smilacina, on the other hand, is correctly characterized. 
Moreover, if we do not include this among those names of 
Necker which, Mr. Bentham says, " deserve to be absolutely 
ignored," we may yet find that the law of priority has an- 
other claim on it. In 1763 a much better botanist than 
Necker, namely, Adanson, founded a genus Tovara (essentially 
the same name as Tovaria) on Polygonum Virginia /von, L., 
which is not unlikely to be taken up as a genus ; and the 
name would supersede Necker's by the same rule that Necker's 
supersedes Desfontaines' Smilacina. All things considered, 
then, this is a case for the application of the homely but use- 
ful rule Quicta non moverc ; and much of Mr. Bentham's 



262 REVIEWS. 

pertinent advice may be condensed into this maxim. But 
there remain nice questions to settle with regard to the names 
and extent of- the Liliaceous genus. 

'* The representing the.Greek aspirate by an h was gen- 
erally neglected by early botanists ; but now, ever since 
Do Candolle altered Elichrysum into Helichrysum, modern 
purists have insisted upon inserting the h in all cases ; and 
this has been so far acquiesced in that it is difficult now to 
object to it, though it has the effect of removing so many 
generic names to a distant part of all indexes, alphabetical 
catalogues, etc. Admitting the propriety of adding the as- 
pirate in new names, I had long declined to alter old names 
on this account ; now, however, I find myself compelled to fol- 
low the current." Which is, on the whole, regrettable, espe- 
cially as Alphonse De Candolle would hold out with him. See 
the latter's comment on his Article G6, in which the remark 
is dropped, that " we do not see why we should be more 
rigorous than the Greeks themselves." Oddly enough, these 
same writers who must supply the aspirate to the e omit it 
from the r, and write rachis and raphe, instead of rhachis and 
rhaphe, — which is exasperating to lovers of uniformity. 

It is unnecessary here to cite Mr. Bentham's appropriate 
illustration of the indivisibility of the two-worded name of a 
plant. The proper apprehension of this, and of the para- 
mount ride that no unnecessary new names should be given 
to old plants, will go far to rid the science of a principal 
remaining ambiguity in nomenclature. For it clearly follows 
that when a plant has a rightful name under its proper genus, 
the specific half of it is not to be changed because of any 
earlier specific name under some other genus, to which the 
plant does not belong. 



ON THE SELF-FERTILIZATION OF PLANTS. 263 



HENSLOW ON THE SELF-FERTILIZATION OF PLANTS. 

This paper 1 is elaborate, mostly able as well as ingenious, 
in all respects considerable, and unconvincing*. Its thesis is 
the Darwinian " Nature abhors perpetual self-fertilization," 
read backward. It concludes that, " not only are the majority 
of plants self -fertilizing, but that those which are exclusively 
so propagate abundantly and w'ith extraordinary rapidity, are 
best able to establish themselves in foreign countries, as, be- 
ing quite independent of insects, they run no risk of exter- 
mination on that score ; . . . that, so far from there being 
any necessarily injurious or evil effects resulting from the 
self-fertilization of plants in a state of nature, they have 
proved themselves to be in every w T ay the best fitted to sur- 
vive in the great struggle for life." The hypothesis is also 
advanced " that they are all degraded forms," and that there- 
fore " their ancestral life-history is a longer one than that of 
their more conspicuous and intercrossing relations." We fail 
to see how this follows, except upon the assumption that the 
earliest phaenogamous plants had the most highly organized 
blossoms ; and that would not accord with vegetable palaeon- 
tology. 

Mr. Henslow rejoices that he has one stanch supporter ; 
" for, as has been seen, Mr. T. Meehan has arrived at the 
same conclusion ; " and indeed he builds not a little upon 
facts supplied by Mr. Meehan's observations. He cites the 
latter's " admirable paper, which was reproduced in the 
' Gardner's Chronicle ' for September 11, 1875, and is in fact 
an ' apology ' for self-fertilization." As he then marshals 
twenty reasons for believing particular plants to be normally 
self-fertilizing, and nineteen " chief facts which may be re- 
garded as occurring correlativelv with self-fertilization, some 
being actual causes which directly or indirectly bring it about," 
it would appear that it is no longer self-fertilization, but 

1 On the Self-Fertilization of Plants. By George Henslow ; Transactions 
Linnrean Society, 2 ser., Bot. i. London, 1879. (American Journal of 
Science and Arts, 3 ser., xvii. 4S9.) 



264 REVIEWS. 

rather the existence and raison d'etre of cross-fertilization 
that stands in need of apology, or of explanation. 

He freely concedes that the flowers of many plants, and 
some whole orders, are so constructed that intercrossing is for 
them a necessity ; also that most of those which are believed 
"to be normally self-fertilizing" because they can and do 
fertilize themselves habitually, yet " may in some cases be 
cross-fertilized by insects." It is admitted that the structure 
of the latter is adapted — most variously and wondrously 
adapted — to being fertilized by particular insects. As this 
conies to pass in plants and flowers of the highest organiza- 
tion and greatest specialization, Darwin and his school con- 
clude that this is a most advantageous outcome, and means 
some real good to the species ; that when this is accompanied 
with a loss of self-fertility, it is the loss of something no 
longer useful, something better than self-fertility having taken 
its place. But Mr. Henslow, reading this the other way, 
having determined " that self-fertilization is per se a decided 
advantage," and free from injurious liability, comes to regard 
intercrossing as merely " a compensatory process for the loss 
of self -fertility." 

But how and why did this " compensatory process " come to 
pass ? It is conceived on both sides that flowers were " pri- 
mordially inconspicuous." (To this Henslow adds herma- 
phrodite and self-fertile, but that need not here come into 
account.) Both agree that insects have mainly determined 
their conspicuousness. Darwin says this has been determined 
through natural selection by the survival of the more and 
more conspicuous variations, correlated with their producing 
something srood for the insect of which the coloration was a 
sign, and that the preferential survival of the more showy 
and attractive was a consequence of some benefit of the inter- 
crossing. Henslow propounds the view that insects have de- 
termined the conspicuousness more directly, and not by bene- 
fiting but by irritating the flowers. " These, by being greatly 
stimulated by the repeated visits of insects, tend to become 
hypertrophied. Hence the corolla, enlarges, becomes more 
brightly colored, the nectariferous organs increase the quan- 



ON THE SELF-FERTILIZATION OF PLANTS. 265 

tlty of secretion, and the stamens develop more pollen. Such 
being the case, nourishment is withheld from the pistil, which 
is delayed in its development ; consequently such a flower is 
very generally proterandrous." Mr. Darwin might accept 
this as an ingenious conception of the way the specialization 
comes about, still insisting on the advantage of the resulting 
intercrossing — " or else the thing would hardly come to pass," 
as the poet has it. And Mr. Henslow's hypothesis has to be 
supplemented to account for proterogyny, which is not much 
less common. But Henslow's supposed process works evil 
instead of good, and is therefore utterly anti-Darwinian and 
" dysteleological." For the result is a disturbance of the 
equilibrium and proper correlation between the androecium 
and gynoecium ; and this, carried further, should upon this 
view result in the monoecious and dioecious states. So, ac- 
cordingly, the cross-fertilization which comes into play in the 
case of separated sexes, and in that of self-sterile hermaphro- 
ditism, is not for any good there is in it per se, but because 
it may no better be. And all the elaborate, exquisite, and 
wonderfully various modes of adaptation of flowers to insects 
are only ways of repairing the damages inflicted upon blos- 
soms by insects through their persistent visits ! Did Mr. 
Henslow ever ask himself the question why the sexes are 
separate in animals ? 

The conclusion which Mr. Darwin has helped us to reacli 
is, that intercrossing should be regarded as the aim in nature 
and on the whole most beneficial, and self-fertilization as a 
safeguard against the risks of crossing ; that most hermaphro- 
dite flowers have the advantage of both, the latter for imme- 
diate sureness, the former for ultimate benefit. Upon the new 
view, self-fertilization is the aim and the consummation, and 
cross-fertilization at best a succedaneum. By it insects may 
repair the damage they have caused to blossoms through en- 
dowing them with " the fatal gift of beauty," and stimulating 
their organs of secretion ; and by it the winds may bring 
chance relief to those which, at length abandoned by their 
spoilers, have lost this attractiveness and fallen to the de- 
gradation of unisexuality. For these last, as has already been 



26G REVIEWS. 

stated, are hypothetically regarded as degraded from higher 
iloral types. 

We are bound to glance at some of the considerations 
which arc adduced in support of this thesis. They are multi- 
farious and of unequal value. As has occurred in other cases, 
so here also, the weightiest objections to Mr. Darwin's view 
are those which he has himself brought out, namely, the fact 
that, as tested experimentally under cultivation, while some 
plants are much increased in vigor and fertility by artificial 
intercrossing, others are not sensibly benefited ; and that the 
benefit derived in marked cases is not cumulative, but reaches 
its maximum in two or three generations. And even close 
breeding under cultivation occasionally gives rise to very 
vigorous and fully prolific self-fertile races. Then many 
plants are fully self-fertile in nature, and it is not proved that 
any such have lost or are in the way of losing either fertility 
or vigor through continued inter-breeding. But, before draw- 
ing from this the conclusion that cross-fertilization is of little 
or no account in nature, it should be remembered that bud- 
propagated races are in similar case. Races exist which have 
been propagated only from buds for hundreds of years, with 
seemingly undiminished vigor, and there is no proof that any 
one has succumbed under the process. But for all that we 
do not doubt that sexual reproduction contributes something 
to the wellbeing of the species, besides facilitating its disper- 
sion. Again, no one questions the necessity of fertilization 
by pollen to the production of embryo in the seed ; yet, even 
in this, the necessity is not so imminent but that some em- 
bryos may originate without it. 

In short, the facts brought out by Darwin and others, and 
all the considerations of the present essays, are best har- 
monized by the conception which the former has consistently 
maintained, namely, that an occasional cross suffices to secure 
the benefit of intercrossing, whatever that may be. Nothing 
vet appears which seriously disturbs our conviction that just 
this is what nature generally provides for. 

Mr. Henslow's proposition, " The majority of flowers are 
self-fertile," is doubtless true in the sense that they are capa- 



ON THE SELF-FERTILIZATION OF PLANTS. 267 

ble of self-fertilization, and is not improbable in the sense 
that they " can and do fertilize themselves habitually."' But 
his inference that the majority of flowers, or that any flowers, 
actually propagate for a series of generations by self-fecunda- 
tion, or that a cross if it occur is " exceptional," and of no , 
account, is surely unwarranted by the evidence which he has 
adduced. 

Occasionally the reported facts will not bear scrutiny. 
Gentiana Andrewsii, it is said, never opens at all in America. 
It opens in sunshine in the middle of the day here in New 
England. And while looking at closed flowers we have seen 
a humble-bee emerge from one. We have in this Journal 
shown how it is that self-fertilization is impossible during the 
first three or four days of anthesis, but neatly practicable after- 
wards. It is rash to infer (as on p. 330) that papilionaceous 
flowers which shed their pollen early in proximity to the 
stigma are therefore self-fertilized. In most of the cases ad- 
duced the pollen is not lodged upon the stigma, but upon the 
style below it, and the adaptations for intercrossing, though 
the mechanism be different, are as explicit as in the analogous 
case of Campanula. " Fremont pathetically describes the 
solitary bee that rested on his shoulder at the top of Pike's 
Peak." The pathos is wasted as respects all but this particu- 
lar bee ; for the entomologists find the alpine region of the 
Rocky Mountains to be as well stocked with flying insects as 
are alpine regions in other parts of the world. They do not 
super-abound, but if from the alpine flora we subtract the evi- 
dently entomophilous and the anemophilous blossoms, the re- 
mainder will be nearly nil. And as to the correlation of this 
comparative scarcity of insects with the marked conspicuous- 
ness of blossoms, this is the way the lesson is read by a most 
eminent physiologist : " Even the glowing hue of alpine flow- 
ers is accounted for by the attraction which brighter-colored 
individuals exercise upon the insects, scarce in those heights 
and necessary for fertilization." 

One or two of the author's own observations are perhaps to 
be revised. " Gaura parviflora . . . has no corolla and is 
cleistogamous, in that it is self-fertilizing in bud, as I found 



V 2G8 REVIEWS. 

in specimens growing at Kcw." Were they not imper- 
fectly developed blossoms, perhaps late in the season ? Here 
the flowers open freely, and have rose-colored petals. If he 
will examine fresh specimens of Scrophularia, it will soon be 
clear that his idea of their self-fertilization (p. 371) is a mis- 
take. It is a mere slip in the " Genera Plantarum " through 
which abortive stamens are attributed to the cleistogamous 
flowers of Epiphegus. The authors evidently meant to 
describe the case just as Mr. Henslow found it to be, but used 
a wrong word. 

" Weeds are probably all self-fertilizing or anemophilous. 
A weed is simply an unattractive plant, and possessing no 
feature worthy of cultivation." It may be as difficult to define 
" a weed " as to define " dirt." But, turning to the " Handbook 
of the British Flora," we find, as we expected, that the showy 
Corn Poppy, Cockle, and Larkspur are denominated weeds. 
Why weeds should possess the vigor and gain the predomi- 
nance which they do is a large question, to which other solu- 
tions have been offered than that one which is in this essay 
very plausibly maintained. We cannot take up the toj)ic 
here ; but, without acceding to his general proposition, we are 
much disposed to agree with the author in this essay, as respects 
some of them, that aptitude for self-fertilization may have 
given them the advantage which has determined their wide 
dispersion. 

The insistence upon the importance of self-fertilization is 
what gives this essay its value. As a whole it fortifies the 
proposition, well laid down by Herman Miiller, which Mr. 
Henslow cites: " that, under certain conditions, the facility 
for self-fertilization is most advantageous to a plant, while, 
under other conditions, the inevitableness of cross-fei'tilization 
by the visits of insects is the more advantageous." But this 
is not our author's thesis. It comes to this : the plan of nature 
is either cross-fertilization supplemented by close-fertilization, 
or close-fertilization tempered by cross-fertilization. As re- 
stricted to plants the difference is not wide. Regarded gen- 
erally, the Darwinian axiom is still best sustained. 



PLANT ARCHEOLOGY. 269 



PLANT ARCHEOLOGY. 

Investigations in fossil botany are recondite and tech- 
nical, the materials generally unattractive, and the results 
unintelligible to the popular mind ; but in Count Saporta's 
" Monde des Plantes," l and under his happy exposition, the 
stony desert is made to rejoice and blossom as the Rose. The 
interest which we take in the vegetation of former periods 
is not so much geological as genealogical ; and this interest 
diminishes with the distance from our own time and environ- 
ment. We know nothing of the earliest plants — the begin- 
nings of vegetable even more than of animal life are beyond 
our ken ; no great satisfaction seems obtainable from the 
small acquaintance that has been made with the plants which 
nourished before the carboniferous period. And the botany 
of that age, notwithstanding its wealth of Ferns and its adum- 
brations of next higher types, impresses us as much with the 
sense of strangeness as of wonderful luxuriance. For even 
the fern-impressions, familiar as they may look to the unpro- 
fessional observer, are outlandish. The more the critical stu- 
dent knows of them the less likeness he finds in them, or in 
the coal-vegetation generally, to any species or genera now 
living. 2 But in the vegetation of cretaceous, and still more 
of tertiary times, familiar forms first come to view, and pedi- 
grees may begin to be traced. Questions of ancestry touch 
us more nearly than those of history ; so an enquiry into the 
source and parentage of the plants with which man is asso- 
ciated is more attractive than any question concerning the 
origin of the pristine vegetation of the earth. Moreover, our 

1 Le Monde des Plantes avant V Apparition de VHomme. Par le Comte 
de Saporta. Paris and New York, 1879. (The Nation, Nos. 742 and 
743, September 18 and 25, 1879.) 

2 To those who wish to get a good coup d'ceil of this vegetation from 
authentic records systematically arranged, we recommend the " Atlas to 
the Coal Flora of Pennsylvania and of the Carboniferous Formation 
throughout the United States,'' by Leo Lesquereux, an octavo volume of 
eighty-seven double plates, just issued by the Second Geological Survey 
of Pennsylvania. There is nothing else to be compared with it. 



270 REVIEWS. 

knowledge of the later fossil botany is comparatively full, — 
wonderfully so, considering how very recent this knowledge is, 
— and we are in a condition to apply it hopefully and confi- 
dently to the solution of problems which not long ago seemed 
to be beyond the reach of proper scientific enquiry, namely, to 
the explanation of the actual distribution of the species of 
plants over the earth. For the main data themselves, and for 
the clear exposition of them, we are most largely indebted to 
three men, who happily are still alive and active — Heer, 
Lesquereux, and Saporta. 

The Linnaeus and facile princeps of tertiary botany is 
Oswald Heer, of Zurich, now a septuagenarian, but still in 
harness. His " Recherches sur la Climat et la Vegetation du 
Pays Tertiare," rendered into French by C. T. Gaudin, was 
published nearly twenty years ago. It is a general and com- 
paratively untechnical presentation of a long line of investiga- 
tions, which have since been crowned by his several memoirs 
on arctic phyto-palaeontology, now collected in the five volumes 
of his " Flora Fossilis Arctica." All these volumes, as well as 
others on the Swiss tertiary, have appeared within the last 
ten years, the latest only a year ago. 

Leo Lesquereux, Heer's compatriot, and barely his junior, 
came to the United States fully thirty years ago, drawn hither 
from Neufchatel by Agassiz. The greater part of his re- 
searches relate to the carboniferous flora, and he has recently 
thrown interesting light upon silurian botany, as has Dawson 
of Montreal upon the intermediate devonian. But those 
which at present concern us relate to the cretaceous and the 
tertiary of our own western regions. The most considerable 
of these works are the two notable quarto volumes, entitled 
" Contributions to the Fossil Flora of the Western Territories," 
published by the Geological and Geographical Survey of the 
Territories under Dr. Hayden, upon whom and whose survey 
they reflect high credit. One volume treats of the cretaceous, 
one of the tei'tiary flora. 

Any proper enumeration of authorities upon the fossil 
botany of the later periods should include various other names, 
and especially that of Schimper, of Strassburg, who, like 



PLANT ARCHEOLOGY. 271 

Lesquereux, has divided his life between bryology and fossil 
botany, and whose classical " Traite de Paleontologie Vege- 
tale " is a systematic compendium of what was known of fossil 
plants up to the year 1874. But the volume now under notice 
is by a younger man, Gaston, Comte de Saporta, a Provencal, 
who has for fifteen years or more been investigating the rich 
tertiary deposits of Aix and vicinity, in the delta of the 
Rhone, the results of which have appeared from time to time 
in memoirs, mainly published in the " Annales des Sciences 
Naturelles." Besides these weightier and more technical pub- 
lications, Count Saporta has contributed to the " Revue des 
Deux Mondes " and to " La Nature " subsidiary articles of a 
popular cast and of fine literary as well as scientific finish. 
These, now collected and re-edited, form a part of the volume 
before us, "Le Monde des Plantes avant 1' Apparition de 
rHomnie," which has been published since the commencement 
of the current year. It is the most comprehensive and the 
most attractive, as well as the most recent, exposition of our 
subject, is a very readable book from beginning to end, of 
inviting typography, with abundant illustrations, both of 
woodcuts in the letter-press and intercalated plates. Although 
a popular, it is a truly scientific volume. The clear stream 
of the narrative is hardly at all troubled by the many tech- 
nical terms which unavoidably strew its course, yet without 
obstructing its flow, for the author has the peculiarly French 
gift of happy exposition. As the volume is likely to be 
reproduced in English, let us hope that it may have a trans- 
lator in whose hands it may lose nothing of its clearness, and 
as little as possible of its freshness and spirit. 

To attempt a popular abstract of such a book would be like 
skimming the cream from the cream, and a critical review 
would cover too much or too technical ground. Still, we may 
give some general idea of the contents of the volume. The 
first part and the most discursive portion of the book is 
entitled u Phenomena and Theories." The introductory chap- 
ter discourses upon the introduction of life and the origin of 
the earliest terrestrial organisms ; and the second chapter takes 
up the theory of Evolution or Transformism. We may skip 



272 REVIEWS. 

these chapters, yet without advising the reader to follow the 
example, unless he is already familiar with the topic — now a 
little threadbare — for, as a popular presentation, it is neat 
and sensible, though not profound. It hardly need be said 
that Saporta is an evolutionist, using the term in its general 
sense, and apparently a thorough Darwinian. A vegetable 
palaeontologist who studies the later geological deposits cannot 
be otherwise ; at least, he must needs be a " transformist." 
Saporta concludes that palaeontology, if it does not furnish 
demonstration, yet gives irresistible reasons for a belief in 
evolution. The ground and the nature of this conviction 
appear in his rounded statement, that there is not a tree or 
shrub in Europe, in North America, at the Canaries, in the 
Mediterranean region, the ancestry of which is not recogniz- 
able, more or less distinctly, in a fossil state. This is too 
absolutely stated, no doubt, but the qualifications it may need 
will not invalidate the conclusion. 

The chapter on ancient climates which follows, and forms 
a proper introduction to the second part of the book, is worthy 
of particular attention. It is prefaced by an elementary but 
very graphic exposition of the phenomena and laws of climates 
and their diversities, from the regular succession of equal days 
and nights under the equator to the contrasted condition 
toward the pole of a year composed of a day and a night sea- 
son, separated by a season of twilight ; the change so rapid in 
the high latitudes that, while the summer day at the North 
Cape is two months long, at Spitzbergen seven additional 
degrees of latitude lengthen it to four months. Let the 
author point the contrast between the two extremes, as affect- 
ing man, in his own language, here somewhat exceptionally 
ornate : — 

"II est vrai que dans ce dernier pays [Spitzbergen] le soleil s'&- 
leve au plus de 37 degre"s au-dessus de l'horizon ; il n'envoie que 
des rayons sans chaleur, telum imbelle sine ictu ; il eclaire de sa 
lueur pale une terre glac^e oil frissonnent quelques plantes enseve- 
lies sous les frimas, et qui ne sortent du sommeil qui les tient clix 
mois inertes que pour accomplir hativement leurs fonctions vitales 
et se rendormir de nouveau. Quel tableau, si Ton songe aux forets 






PLANT ARCHEOLOGY. 273 

vierges clu Bre"sil et de Java, aux valines profondes du Nepaul, aux 
savanes noyees de l'Orenoque, ou la vie surabonde, ou une lumiere 
ardente, vive et doree, ondule de toutes parts, souleve de tiedes 
vapeurs, joue avec l'ombre, et fait resplendir les formes des plus 
merveilleux vege'taux! Sous les tropiques, l'homme se sent ecrase" 
par une vie exuberante, il lutte incessament pour maintenir sa place 
au milieu de la nature, dout il est domine" ; ses plus fortes eeuvres sont 
envahies en peu de temps ; les arbres immenses reprennent possession 
du sol, des que celui-ci est abandonne" a lui-meme. Dans l'extreme 
Nord, la faiblesse de l'homme est encore plus eVidente, mais c'est 
du poids de la nature inerte qu'il est accable". Les elements regnent 
seuls dans ces regions deVastees, oil l'atmospliere se trouve livree a 
d'epouvan tables tourmentes. La neige derobe les asperites du sol, la 
glace couvre la mer d'un sol factice, souvent mobile et toujours dan- 
gereux ; la confusion est partout, le calme nulle part ; cbaque pas est 
penible, la vie elle-meme devient un effort que l'energie la mieux 
trempee ne peut soutenir longtemps sans succomber." (p. 212.) 

The modifications of this contrast through the actual dis- 
tribution of land and water, winds and currents, are then con- 
sidered. As these have a fixity secondary only to the funda- 
mental elements of climate, — namely, the heat of the sun, the 
inclination of the earth's axis to the plane of its orbit, and 
the relative density of the atmosphere according to the eleva- 
tion of the land-surface, — the climate of any part of the world 
might be supposed to have been constant, oscillations excepted, 
through the long periods that have elapsed since existing 
species or their immediate ancestors were introduced. It is 
not very long ago that Arago demonstrated, to his own and 
the general satisfaction, that there has been no appreciable 
change of the earth's climate iu man's time. Plants are the 
thermometers of the ages, by which climatic extremes and 
climates in general through long periods are best measured. 
For at least five or six thousand years the Vine and the Date- 
palm have grown in proximity, and have furnished grapes 
and dates to the inhabitants of the warmer shores of the 
Mediterranean. Yet a very moderate change either way in 
the temperature would have excluded the one or the other. 
So Arago concluded that man had witnessed no sensible 
changes in the climate of Europe ; a good conclusion, if re- 



274 REVIEWS. 

stricted within the limits of observation. But in Arago's life- 
time the evidence was already accumulating which has now 
proved that in earlier times man and the reindeer lived 
together on the soil of southern France, when if grapes could 
ripen in Syria and northern Egypt, dates doubtless could not. 
Then, at a still earlier day, Palms flourished in Switzerland 
and Vines in Iceland. Then Maples, Lindens, Plane-trees, 
Spruces, and Pines formed forests in Greenland uji at least to 
the eightieth parallel ; and, indeed, our own southern Cypress, 
or Taxodium — which now barely maintains its existence 
between the mouths of Delaware Bay and the Chesapeake — 
flourished along with the Silver Fir of Europe, within two 
hundred leagues of the North Pole. A climate in Greenland 
in which Sequoias, now confined to California, Magnolias, 
Persimmons, and Grapevines were mixed with Maples, Oaks, 
and Poplars, could not have been colder than, or much unlike, 
that of Indiana and Kentucky now. 

As w r e have seen, the vegetable world has had an eventful 
history, and this history Count Saporta undertakes to recon- 
struct from ancient and authentic documents. 

These documents have settled one point with certainty, 
namely, that the great changes in the temperature of the polar 
regions have not resulted from any change in the earth's axis 
of rotation, such as certain physical geologists have supposed. 
That this has remained steady throughout the whole period in 
question is as good as proved by the identity of the miocene 
and other tertiary fossil plants in all longitudes, from the 
Mackenzie River and Alaska round to Spitzbergen, Iceland, 
and Greenland — in part the same species, in all the same 
or equivalent combinations. Not indeed the same species in 
the same latitude any more than now, but such latitudinal 
distribution as to show that the curvatures of the miocene 
isotherms were quite analogous to those of the present age. 
Moreover, the monotony which characterizes the sub-polar 
vegetation of the present day — when the most of the species 
and the combinations of species occur all round the world — 
equally characterized it then when clothed with forest trees of 
a temperate zone. 



PLANT ARCHEOLOGY. 275 

Upon the data in hand, now much extended, and on the 
supposition that the same species of tree had not appreciably 
altered meanwhile in its relations to temperature, Heer long 
ago elaborately compared the miocene climates with those of 
our time, and Saporta corroborates his conclusions. For the 
northern regions the difference is said to be equivalent to 25 
or 30 degrees of latitude — that is, we have now, say in Eu- 
rope, in latitude 40° and 45°, and in Atlantic America, in 
latitude 38° to 40°, the temperature and the vegetation which 
then flourished at latitude 70° in Greenland. Grinnell Land, 
in latitude 82°, only two hundred leagues from the pole, had 
a forest of coniferous trees ; among them — associated with a 
Poplar-tree, a Hazel, and a Birch — was the Silver Fir of 
Europe, and the Bald Cypress of the swamps of the southern 
United States. The same combination, minus the Cypress, 
must now be sought in the more elevated parts of central and 
southern Germany. The Sequoias and Magnolias and Per- 
simmons of Greenland, in latitude 70°, mingled with Maples, 
Oaks, and Grapevines, have their representatives partly in 
Virginia and on the Ohio River, partly in California. The 
miocene of the southern shore of the Baltic had Laurels, Ole- 
anders, and Camphor-trees, but no Palms, so far as is known, 
thus answering to the Mediterranean flora, but miocene Palms 
reached to Belgium and Bohemia ; and Provence, on similar 
data, had then about the climate of the coast of Zanzibar. 

Another conclusion which Saporta confidently reaches — 
and which indeed is reached from all sides — is that of a very 
moist quaternary climate. Looking back to this compara- 
tively recent period from our own, everywhere the streams 
have dwindled. Through great river-beds shrunken stream- 
lets now meander in insignificant channels ; springs reach the 
surface much lower down the valleys than of old ; and the 
" rivers without water " of Egypt and Syria, and the reduced 
level of the Dead Sea, are so many evidences of a dryness 
supervening upon a general humidity greatly in excess of the 
present. These fuller watercourses of themselves indicate a 
more temperate or mean climate, a more equal distribution of 
heat and cold through the year. Was this equable climate 



276 REVIEWS. 

frigid and rigorous or comparatively mild ? Were Europe 
and the United States simply arctic and to be compared with 
the present Greenland and Spitzbergen, as the school of 
Agassiz maintains? Or should the comparison rather be 
made with southern New Zealand, where Tree-ferns almost 
overhang the terminal moraines of existing glaciers, in a cli- 
mate which is neither cold nor warm? Saporta maintains 
the latter, and he is not alone. He insists that the high Alps 
and the Pyrenees are not the types of glacial Europe gener- 
ally ; that the arctic animals and plants, and the rigorous cli- 
mate which we associate with these, belonged only to the close 
neighborhood of glaciers, but that the valleys below enjoyed 
a climate even milder than now, although vastly more humid. 
So, likewise, Mr. Ball (in a lecture recently delivered before 
the Royal Geographical Society of Great Britain) ventures to 
affirm that, even during the pei'iod of maximum cold, the 
highest ridges of the Alps were not completely covered with 
snow and ice ; for we still see, by the appearance of the sur- 
face, the limit above which the ancient ice did not reach ; and 
in the middle zone the slopes that rose above the ancient gla- 
ciers had a summer climate not very different from that which 
now prevails. And he concludes that the effect on the growth 
of plants in the Alps was to lower the vertical height of the 
zones of vegetation only one or two thousand feet. 1 

This would seriously affect the forests of Europe, but would 
not permanently disturb the alpine and sub-alpine vegetation. 

Yet cold it must have been when the reindeer and musk-ox 
roamed over the plains of central Europe, and when the ele- 
phant or mammoth and even the rhinoceros which accom- 
panied them, were equally clad with a thick coat of hair. 
But, says Saporta, with the remains of these very animals 
from which a frigid arctic climate is inferred, occur also, in 
the alluvia of the Somme and the Seine, those of an elephant 
nearly related to the Indian species, the hippopotamus of the 
African rivers, and the hyena of the Cape ; and the vegetable 

1 But Mr. Ball is fairlv astounding: when he assumes that our arctic- 
alpine flora may have been the flora of high mountains at low latitudes in 
the carboniferous period. 



PLANT ARCHEOLOGY. 277 

remains include the Laurel of the Canaries along with the 
Vine. The trees of the same epoch farther north were Pines, 
Lindens, Maples, and Oaks. So, according to Saporta, even 
the glacial period formed only a seeming interruption to the 
general course, the steady and really unbroken diminution of 
terrestrial temperature from the earliest geological periods to 
the present. This must be admitted if the two classes of ani- 
mals and plants — those adapted to cold and those to warmer 
climates — were really contemporaneous. Our geologists have 
maintained that they were not, but that climates have oscil- 
lated, and that warmer periods than ours intervened between 
the glacial epoch and the present, or were intercalated in the 
glacial period itself. But is not the distinction of periods an 
assumption for explaining the two kinds of fossil remains ? 

We need not enter here into the discussion of the cause of 
the higher temperature of ancient climates, and of that pecu- 
liar and temporary state of things attending and originating 
the glacial epoch, with which Saporta concludes his third 
chapter. And no space is left us in which to sketch even the 
outlines of the second part and main staple of his book, the 
history of the vegetable periods, beginning with the " primor- 
dial marine plants " of the Laurentian and closing with the 
pliocene, in which existing trees are everywhere identified. 
The general conclusion of these very rich, elaborate, and well- 
considered chapters is that the vegetation of the earth has 
been continuous through all ages, and that the explanation of 
the present is found in the past. The history of the genus 
Sequoia — of the two " big trees of California " — as recently 
sketched by Heer in a popular journal, " Das Ausland," is a 
fair illustration of this. The difference between these two 
trees is as notable as their resemblance and their isolation. 
They are the survivors of a numerous family, of wide dis- 
tribution, which is first recognized in the cretaceous forma- 
tion, in several species, and which reached its maximum in 
the middle tertiary, in fourteen recognizable species or forms. 
Almost from the first these separate into two groups, one fore- 
shadowing the Coast, the other the Sierra, Redwood, yet with 
various intermediate forms. These intermediate species are 



278 REVIEWS. 

extinct, the two extreme forms have survived. The likeness 
of these two trees is explained by their genealogy, their 
marked difference by the extinction of the connecting forms 
which in earlier times bridge the interval. 



SERENO WATSON ON NORTH AMERICAN LILIACE^. 1 

Mr. Watson, in preparing the Monocotyledoneee for the 
" Botany of California," came upon the order Liliacece, which 
is well represented in Pacific North America ; and he had 
to consider how the genera and higher groups should be dis- 
posed. This led to a wide study of the order and a strict 
scrutiny of the American species ; and the present " Revision 
of the North American Liliaceoe," occupying the greater part 
of the " Contribution " before us, is the result. It is gen- 
erally agreed that this order is to have the wide extension 
which was given to it by the present writer a dozen and more 
years ago ; and the proper collocation of its diversified forms, 
with interlaced affinities, has been a problem of no small diffi- 
culty. Mr. Baker, in England, has attempted the task for the 
order generally, and has sedulously elaborated some of the 
North American, but more of the Old World and the South 
American, genera and tribes. His arrangement and his sys- 
tematic views are in many respects satisfactory, in some un- 
satisfactory as respects North American botany. Mr. Watson 
has the latter primarily in view, but still has to adjust the 
American genera into the general system. The arrangement 
he has planned consists of three series, the first of which parts 
into two subseries, and includes sixteen tribes, some of them 
divided into subtribes. The great endeavor has evidently 
been to make natural groups — and this endeavor has been 
really successful. The next thing is to assign characters, and 
here comes the difficulty. Absolute characters of the lead- 

1 Contributions to American Botany, IX. By Sereno Watson ; Proc. 
American Academy of Arts and Sciences. Boston, 1879. (American 
Journal of Science and Arts, 3 ser., xviii. 313.} 



WATSON ON NORTH AMERICAN LILJACE.E. 279 

ing groups are not to be had, even when North American 
forms only are considered. Those who imagine they could do 
better than Mr. Watson has done should make trial before 
they criticise. The character of the pericarp, whether bac- 
cate or capsular, the nature of the stock, whether bulbous, 
tuberous, or rhizomatous, the nature of the seed-coat, the 
inflorescence, direction of anthers, union or separation of 
styles, are all good characters to a certain extent, and all fail 
to furnish unexceptionable marks to distinguish the higher 
groups when natural associations are sought. It is not easy 
to ascertain what diagnostic characters in this monograph are 
most to be trusted. But the nature of the bracts (on the one 
hand scarious, on the other foliaceous or none) takes the lead 
in the first two series, and is followed by the persistence or 
deciduousness of the perianth, the insertion of the stamens 
whether on the perianth or at its base, the dehiscence of cap- 
sule, — all matters of little physiological importance, but for 
that reason perhaps surer guides to affinity than the more 
prominent adaptive characters. However, it may be said that 
the first series answers to the Asj^hodelece, with Yucca and 
HemerocallidecB added ; the second to the true Liliacece, with 
Uvulariece, and TriUiecc added ; the third to Mdanthaceoe, 
with the tribe Tofieldieoi appended. Thus disposed, it is 
doubtless judicious to designate the three primary groups as 
" series," and not as suborders, and to throw the stress upon 
the tribes. 

The Melanthaceous series, which in our view best divides 
into the Colchicece, Veratrece, and Tojichliece, — the first not 
American, — is here divided into the Veratrece, Heloniem, and 
XerophyllecB (which two we should combine), into the midst 
of which the Tqfieldiece are intercalated. This last tribe, 
which should end the series, is quite exceptional, and is well 
composed of Tofieldia, Pleea, and Narthecium. Its marks are 
the equitant leaves, introrse anthers with parallel cells, and 
caudate seeds ; but to bring Narthecium under the remain- 
ing character of " styles distinct or none," it is defined as 
destitute of style, but with " the slightly lobed stigma sessile 
upon the attenuated apex of the ovary." This is really much 



2S0 REVIEWS. 

nearer the fact than would be supposed, as the cells of the 
ovary actually do taper up into the subulate style (as it has 
always and most naturally been termed), so that in the ma- 
ture capsule the upper tails of the seeds reach up to within a 
short distance of the small stigma. 

In a linear order it has not been practicable to approxi- 
mate the ConvcUlariece of the first series with the Uvulariem 
of the second. The division of Uvularia gives a gratifying 
opportunity of dedicating a New England genus to the mem- 
ory of one of the best of New England botanists, the late 
William Oakes (Oakesia sessilifolia, with its relative of the 
southern mountains, O. puberula) ; but he would not have 
relished the dismemberment of the Linnasan genus upon the 
characters, good as they are, neither in fact do we. The 
formation of the tribe Yuccece, of Yucca and Hesj)eraloe, 
strikes us as excellent ; and it seems right not to adopt the 
supposed second species of Hesperaloe until it is better 
known. Its principal distinctions (longer anthers and shorter 
st}de) may indicate heterogone dimorphism, which would be 
a novelty in the order. 

Tribe JVolinece, of the first series, must be regarded as an 
excellent group, composed of Dasylirion and Nolina ; and it 
is gratifying to find that the outlying genus Nolina, founded 
long ago on a single Georgian species, is the northern repre- 
sentative of a considerable Texano-Mexican group, named 
Beaucarnea. It were to be wished that the plan of this Re- 
vision had allowed more citation of generic synonymy, and 
that it had been more explicitly stated that Beaucarnea is only 
Nolina. This union, indeed, is one of the happy hits of the 
present monograph. 

As has been suspected, the Calif ornian /Schoenolirion album 
of Durand proves to be quite distinct from the Atlantic 
species on which that genus was founded. So Mr. Watson 
has embraced the opportunity, here offered, to dedicate a 
peculiar Californian genus to Judge Hastings — a judicious 
patron both of botanical and legal learning. Except for his 
exertions, his own liberality, and his direction of the liberality 
of others, we could not have had the " Botany of California," 



WATSON ON NORTH AMERICAN L1LIACEJE. 281 

which Mr. Watson may now soon bring to a completion. 
The reader finds no mention of this under the genus Hast- 
ingsia, p. 217, nor under the species H. alba, p. 242. But 
an appropriate reference is made on p. 286. 

Leucocrinum, Nutt., was conjectured by Endlicher to be 
the Mexican Weldenia, but it has just now been ascertained 
at Kew that Weldenia is a Commelynaceous genus. 

Our species of Allium as now woi'ked out by Mr. Watson 
with great painstaking, are thirty-six in number, exclusive of 
the introduced A. vhieale. Some characters might be made 
more of in living plants, such especially as those furnished by 
the so-called " crests of the ovary." In A. stellatum these 
crests are remarkably developed, radiating from around the 
base of the style and recurving, the notch at the end of each 
fitting over the base of the alternate filaments, and the under 
side is nectariferous and attractive to bees. The flowers are 
proterandrous. 

In separating the two species of Maianthemum we should 
have unhesitatingly referred the large Pacific coast form to 
M. bifolium. We should not have distinguished Lillium 
Grayi as more than a form of L. Canadense, one which 
extends northward to the central parts of New York. In 
view of geographical range, size, and general appearance, we 
should never have thought of Uvularia flava as a synonym 
of U. grandiflora. Mr. Watson finds good characters in the 
shape and markings of the capsule to separate U. grandiflora 
from U. perfoliata. Has any one ripe fruit of the small, 
yellow-flowered U. flava ? 

Chamailirium Carolinanum, Willd. This specific name 
is properly restored. It was the original name under this 
genus ; and the name luteum is a false one (though the plant 
was Veratrum luteum of Linnssus), the blossoms being white 
without a tinge of yellow, duller white in the female plant, 
pure white in the male, the pedicels equally of this color. 

No space is left in which to notice the Notes upon the 
Affinities and Geographical Distribution of LiliacecB, nor the 
Descriptions of some New Species of North American Plants, 
about fifty in number, which make up the second part of 



282 REVIEWS. 

this important " Contribution." Among them is a new 
Bolandra and a new Snllivantia from Oregon, both very 
much like (we fear too like) the original species. Here and 
in the Bibliographical Index, the name Sullivuntia Ohionis 
is changed (perhaps accidentally) to S. Ohioensis. We 
knew of no law against genetive names of geographical more 
than of other places or stations, and such are not extremely 
uncommon. The name Ohionis was purposely chosen, and 
we hope may be retained. 

The interesting new Erigoneous genus Hollisteria, discov- 
ered by the enthusiastic Mr. Lemmon (in San Luis Obispo 
County, east of the Coast Range), is of rather doubtful in- 
terpretation as to some points of structure. The inflorescence 
we suppose to be only seemingly axillary, the involucre is 
possibly a genuine trimerous one, and we take the two small 
stipule-like leaves to be real stipules, — a point which the 
published character does not decide, though it is implied in 
describing the leaves as alternate. 

Being one of the most important of recent contributions to 
Xorth American Botany, this publication deserves even a 
fuller notice than we can here give it. 



DE CANDOLLE'S PHYTOGRAPHY. 

Valuable as the present volume * is, it may very probably 
not be translated into English. So we propose to give a run- 
ning account of its contents, adding here and there some brief 
comments, critical or otherwise. Treatises like this can be 
written only by botanists of long experience ; and long ex- 
perience, founded upon good training and accompanied by 
good judgment, gives the right to speak with a certain author- 
ity, particularly upon writing and publication in systematic 
botany, in which rules and method are most important. De 

1 La Phytographie, ou V Art de deer ire les Ve'ge'taux consider es sous 
differents points de vice. Par Alphonse De Candolle. Paris, 1880. 
(American Journal of Science and Arts, 3 ser., xx. 150, 241.) 



DE CANDOLLE' S PHYTOGRAPHY. 283 

Candolle is now one of our oldest systematists, one who as 
editor as well as author has had to consider every sort of phy- 
tographical question ; and the volume he has hex-e produced is 
a needful supplement to the " Philosophia Botanica " of Lin- 
naeus and the " Theorie Elementaire " of the elder De Can- 
dolle, the two classical books which the serious botanical stu- 
dent should early and thoroughly master. Phytography has 
to do with form and method in botanical works ; and natural 
history is nothing if not methodical. Its advancement by 
research and its educational value — which will be more and 
more appreciated as it is better taught — both depend upon 
correct morphology and upon well-settled method. Those 
who will not use its proper language and respect its customs, 
must not expect to be listened to, any more than is unavoid- 
able. Observation and interpretation must go together, if 
either is to be of value ; the naturalist must not only observe 
that he may describe, but describe if he would observe. In 
his preface De Candolle remarks upon the peculiar advantage 
of natural-history study in the combination of observation 
with judgment, and upon the importance to a student of 
acquiring a clear idea of what natural groups are, what a 
natural classification and the subordination of groups really 
mean, and how a naturalist arranges, names, and with pre- 
cision defines the immensely numerous objects of his study. 
Men who have distinguished themselves in various professions 
and lines of life, have owed the advantage they have derived 
from this kind of training in youth, even though they never 
became naturalists. 

De Candolle's book is in thirty chapters, many of them 
short and somewhat discursive, and generally abounding in 
recommendation and advice, rather than laying down positive 
rules. 

The first chapter glances at " the evolution of botanical 
works " from Cesalpino, with whom scientific botany began 
in the middle of the sixteenth century, to Linnaeus, whose rules 
and spirit still govern, and to our own times, noting the grad- 
ually increasing importance of herbaria as compared with 
botanic gardens. The second chapter touches upon the moral 



284 REVIEWS. 

and intellectual dispositions necessary in botanical work, and 
asks the question what manner of men botanists are or ought 
to be. As their pursuits do not lead to fortune, and profes- 
sorships are neither numerous nor well paid, he concludes 
that botany is just the science for disinterested people to 
prosecute from pure love of knowledge and the pleasure of 
discovering something new ; that it does not deal with ques- 
tions of a very high order, nor require very difficult or abso- 
lutely rigorous reasoning. The faculties which it brings into 
requisition are the spirit of observation and of order, sagacity, 
and a certain good sense in the appreciation of facts ; that, if 
it does not shine with great eclat, at least the faults of its 
cultivators are not likely to harm any one ; that, equally with 
the other sciences, it tends to elevation of character in that it 
requires an ardent love of truth, reposing as it does upon the 
idea that the veracity of its cultivators is absolutely complete. 
He concludes : " Les sciences jouent dans le monde le role 
d'une ecole practique de bonne foi. D'apres ces reflexions, 
il est permis de penser que les botanistes sont ordinairement 
et devraient etre toujours des hommes paisibles, inoffensifs, 
indulgent pour les erreurs de leur confreres, et occupes bien 
plus de l'avancement de la science que de leurs interets ou de 
leurs petites glorioles. . . . Ne nous arretons pas cependant 
sur de rares exceptions. La presque totalite des botanistes 
est penetree du sentiment de la justice et des convenances. 
On en trouverait difficilement un seul qui ne reconniit le 
principe fondamental de ne pas faire a autrui ce qu'on ne 
voudrait j^as qui vous fut fait." 

Still, our author continues, sometimes the perfectly honest 
and right-minded botanist may have failings. He may, for 
example, neglect to cite his predecessors, or cite them inex- 
actly, either from negligence (not to speak of calculated omis- 
sions, which show want of honesty and soon bring down 
reprobation), or from the want of literary resources. The 
latter case may be deemed a misfortune, and no fault. But, 
our author rejoins, if he has not the necessary books within 
his reach, why not go where they are and consult them ? Or 
if unable to do that, why need he publish ? 



BE CANDOLLE'S PHYTOGRAPHY. 285 

Some good advice follows about polemics and captious crit- 
icism ; which we pass over, as seemingly superfluous, so long 
as the botanists are almost without exception such peaceable 
and good people. Something is said of the need of a right 
appreciation of the extent of the science ; of the danger of 
exclusive devotion to a single branch of botany, in which one 
may lose all just perspective ; and, finally, of what accuracy 
means in natural history as distinguished from mathematical 
exactness. Everywhere the naturalist has to judge as well as 
to measure. 

The third chapter discourses upon the manner of preparing 
and editing botanical works, and the most advantageous modes 
of publication, considers the different degrees of publicity : 
for instance, complete and durable publicity is attained when 
a Monograph of an order or genus, a Flora, a Species, or a 
Genera Plantarum is published and placed on sale by the 
booksellers ; or when an article or memoir is contributed to 
any leading and well-known botanical journal, or to the bulle- 
tin of a purely botanical society, which publishes with some 
regularity and indexes its volumes ; or when printed in the 
transactions or bulletins of any scientific society, if separate 
copies in sufficient number are printed and fairly distributed 
or placed on sale. The usage in some learned societies of 
paging each memoir separately and placing it separately on 
sale is referred to, with implied commendation. Let us add 
that in all such separate issues, the original pagination of the 
volume should be scrupulously preserved ; and it were better 
that there should be no other. Less complete, but durable 
publication is that of ouvrages de luxe, so limited in number 
of copies, and so high in price, that only a few libraries can 
possess them ; also articles in journals without full indexes, 
or with indexes only to a series of volumes. 

Incomplete publicity is given when papers upon botany are 
inserted in the voluminous transactions of general learned 
societies, of which few individuals can possess the series or 
find room for them ; also articles in reviews, encyclopaedias, 
and the like, treating of many or of all sciences. Even jour- 
nals of natural history alone fall under the ban, unless divided 



286 REVIEWS. 

into separate parts for zoology and botany, — as is the long- 
continued " Annales des Sciences Naturelles " of Paris, and the 
" Journal of the Linnaean Society of London." A remedy or 
alleviation of these obstacles to publicity, and of others like 
them, is supplied by the catalogue of papers published by the 
Royal Society of London ; which noble work was instigated 
by the late Professor Henry ; yet, as this embraces all sciences 
and fills a goodly series of volumes, it can seldom be in the 
library of botanists. 

In speaking of the obstacles to scientific publicity which 
are interposed by too limited editions, high prices caused by 
undue luxury in plates, and inopportune or inappropriate 
media of publication, De Candolle refers to customs in the 
book trade and in government patronage which need reform ; 
and mentions incidentally what a botanical library costs. He 
says there should be by the side of every great herbarium and 
every considerable botanic garden, a special botanical library, 
without which it is impossible to determine exactly the plants 
of the one or the other, or to write any good Monograph or 
Flora. Such a library costs fifty or sixty thousand francs (ten 
or twelve thousand dollars), and needs about 12,000 francs 
for annual purchases. He asks how many such establishments 
there are in the world, and concludes that there may perhaps 
be between ten and twenty. 

The section on the comparative superiority of certain kinds 
of works, sets forth the greater value of books or systematic 
works as compared with memoirs or articles. 

The language to be employed in botanical publications is 
the topic of a special article. For descriptions, Latin, and the 
Latin of Linnaeus. " Le Latin des botanistes n'est pas cette 
langue obscure et a reticences de Tacite, obscure et a pe"- 
riodes pompeuses de Ciceron, obscure et a graces tortille'es 
d'Horace, qu'on nous fait apprendre au college. Ce n'est pas 
meme le langage plus sobre et plus clair d'un naturaliste tel 
que Pline. C'est le Latin arrange par Linne a l'usage des 
descriptions et, j'oserai dire, a l'usage de ceux qui n'aiment 
ni les complications grammaticales, ni les phrases disposees 
sens dessus dessous, ni les parentheses enchassees dans les 
phrases." 



DE CANDOLLE' S PHYTOGRAPHY. 287 

This for descriptions, except in local Floras, where popular 
use demands the vernacular ; and we interpose the remark 
that English botanical language, freely incorporating as it 
does all Latin and Greek terms, comes next to Latin in con- 
venience, compactness, and facility to all foreign botanists, 
who, being familiar with Latin, can seldom be at a loss. For 
discussions and reasonings, the botanists of each nation prefer 
their vernacular tongue ; but De Candolle would restrict them 
to the four modern languages, one or two of which, beside his 
native tongue, every naturalist is nowadays supposed to be 
able fairly to read ; English, German, French, and Italian. 
Indeed, De Candolle recommends Latin and the technically 
descriptive style even for generalia, on the ground of brevity ; 
and he aptly suggests that the less capable botanists are of 
handling other than Liunsean Latin, the more brief, senten- 
tious, and strictly to the point their exposition will be. 

Hints are given as to the best mode of collecting literary 
material, making and preserving notes (each upon separate 
slips of paper) ; upon the importance of adding clear explana- 
tions of drawings at the time they are made ; and upon the 
desirability of refraining from publication until the work is 
thoroughly completed, but of then publishing as soon as pos- 
sible. A manuscript work is said to have its maximum value 
at the moment of completion. Our author declares that sec- 
ond and third editions are seldom equal to the first. That 
depends. He objects also to posthumous publication, citing 
^Roxburgh's " Flora Indica," published by Wallich, Plunder's 
plates, published by Burmann, and the wretched figures of 
Velloso ; and he might have referred to the ill-advised print- 
ing of Griffith's rough notes and comments. But all depends 
upon the character of the manuscript and the length of time 
which has elapsed. 

Chapters IV-XI traverse the whole subject of descriptions, 
under various aspects and a rather minute division of topics. 
As even a brief analysis would overpass available space, we 
will merely touch here and there upon certain points. 

As to the relation of varieties to species, there are two 
modes of presentation, both of which have been followed by 



288 REVIEWS. 

Linnaeus, and by most systematists, upon different occasions. 
Varieties are commonly designated by the small letters of the 
Greek alphabet, a, (3, y, etc., and also by names when they are 
pretty distinctly marked. Either the varieties, one or more, 
may be appended to the species (that is, to the form taken as 
the type, which usually must be the form originally described 
under the name), and therefore be treated as aberrant forms; 
or else the species is characterized as a group of forms, which 
forms are classified and defined just as species are under their 
genus. For instance, Mentha Canadensis is held to compre- 
hend both a hairy and a smooth form, the two differing also 
somewhat in other respects. Linnaeus founded the species on 
the former ; and it is pretty well agreed that we are to refer 
the species back to him, however it be limited. Now we 
may either give a common character to the species, and then 
distinguish var. a. villosa, and var. /3. glabrata ; or we may 
characterize the species in general upon the originally named 
form, and append the variety /3. glabrata. Either mode has 
its advantages and is likely to be employed in certain cases. 
The former classifies the varieties under the species, perhaps 
more naturally, and exhibits the polymorphous character of 
what we call a variable species ; and De Candolle considers 
that it will prevail in proportion as the forms of a species 
come to be well known ; the latter holds closer to the bibli- 
ography. There is danger of some misunderstanding when 
the two modes are used in the same work. In the " Synoptical 
Flora of North America," the former mode is invariably 
adopted, partly on the score of brevity. Either the originally 
described form, or a medium or common form is taken as the 
type, and the varieties are treated as departures from this. 
Even when the specific character is drawn so as generally to 
cover the varieties (as should be done as far as possible), 
some form, and the history of the species generally indicates 
what form, is kept in view as the norm or alpha. Of course, 
except for cultivated plants, there is no knowing and no pre- 
tence of determining which was the parent form, or in what 
order the several varieties may have diverged from a pristine 
stock. 



DE CANDOLLE'S PHYTOGRAPHY. 289 

As De Candolle points out, there is much ambiguity and 
looseness in the use of this word " type," which it would be 
well to avoid. Properly the type of a species is the species or 
genus, or the full idea of it, which no one individual or species 
may embody, which in the case of a group no single represen- 
tative or member can fully exemplify. To apply the term to 
a form which well exemplifies the essential characters of the 
species or genus is quite natural, and hardly involves any 
confusion. But the term is also used in a historical sense, 
as referring to the particular form on which a species was 
founded, or the species on which the genus was characterized 
or which its founder had mainly in view, but which very often 
proves not to be the best representative of the group, some- 
times not even a fair one. Finally a particular specimen 
which the original author described, or an authentic specimen, 
is said to be a type, or a typical specimen ; and this De Candolle 
objects to. But after all, such terms can hardly he held to a 
single sense in technical any more than in ordinary language. 
Something must be left for the context to determine. 

In drawing up the characters of groups, such especially as 
orders and genera are exceptions or what we call exceptions to 
be indicated in the character ; or shall this express only what 
is generally true ? De Candolle discusses the question, but 
leaves it, as must needs be, for practical judgment to determine. 
On the one hand the point or the usefulness of a character is 
blunted or dissipated by the intercalation of alternatives and 
exceptions, yet characters must be somehow made to corre- 
spond with the facts. The method of Bentham and Hooker, 
of a separate specification of the principal known exceptions, 
is commended. 

Should outlying or anomalous groups be incorporated with 
the orders they most resemble, or be merely appended as 
" genera amnia," and the like ? The latter was inevitable in 
the earlier days of the natural system ; but increasing knowl- 
edge, as well as considerations of symmetry and convenience, 
more and more fix the place of these floating groups ; so that 
their general incorporation into the orders by Bentham and 
Hooker in the Genera Plantarum of our day is in the natural 



200 REVIEWS. 

course of things. But botanists have to remember that many 
of them are still riddles. 

De Candolle classes descriptions under the two general heads 
of developed and abridged. A developed description is a 
detailed account of the whole conformation, without regard 
to differentia?. The type of an abridged description is the 
diagnosis, such as the specific phrase, or as Linnaeus called it, 
the nomen xpec'ijicum ; what we now universally term the 
specific name being his nomen trivicde. In the course of 
phytography both these have become rare or of special use 
as regards species, and a hybrid between the two has been 
engendered which is more serviceable than either. The long 
and independent descriptions of the olden time are now seldom 
written. Except for special cases, the development of the 
natural system in its subordination of groups in ever increas- 
ing numbers and definiteness, has rendered them superfluous. 
What was once stated in the developed description of a species 
in one formula, and a vast deal more, is now parceled out 
among the ordinal, tribal, generic, sub-generic, or sectional, 
sub-sectional, and other characters, each of which deals pri- 
marily, if not wholly, with differentiae. The characters of each 
grade, being diagnostic, may be comparatively short ; but 
taken together they become almost exhaustive. But to avoid 
going again over the same ground, subsidiary matters not 
diagnostical, yet needful or useful, are not rarely intercalated 
among the more essential points, instead of being collected in 
a separate paragraph. Consequently the specific diagnosis 
may be prolonged and get to partake of the nature of a 
developed description. The remedy for over-length is to 
multiply divisions and sub-divisions between the genus and 
the species. To do this well, to arrange the species group 
within group most definably as well as most naturally, tasks 
the powers and the patience of a systematic botanist, and tests 
his aptitude for discerning affinities, and solving practical 
difficulties. 

Developed descriptions are in place in such general works as 
De Candolle's " Systema " (which was soon overweighted and 
crushed by them), and above all in monographs of orders or 



BE CANDOLLE'S PHYTOGRAPHY. 291 

genera. In his sixth chapter, devoted to this topic, the author 
cites, in order of date, the principal monographs of orders or 
tribes (excluding those of a single genus) which may be taken 
as models (about two dozen only), and points out some of 
their merits or defects. The subject of abridged descriptions 
is taken up in chapter VII. ; and this connects itself with a 
great number of subsidiary questions and particular details, 
running on through twenty chapters more (individually short), 
and forming the most practically useful part of the book. 
There are so many points which it were well to call attention 
to, or sometimes to comment upon, for which space is now 
wanting, that we must defer the remainder of this critical notice 
to the next issue of the Journal. 

It is to be regretted that, for the completeness of this work, 
the author did not comprehend in it the subject of nomen- 
clature of groups — an important part of phytography — and 
reprint in it his opuscula, entitled " Lois de la Nomenclature 
Botanique," along with some further commentaries, such as 
his experience and some adverse criticisms from an opposing 
school may have suggested. This may still be desired, 
although the little treatise has already been widely dissemi- 
nated in three languages, and although, as the author inci- 
dentally remarks, his own view is shared by an immense 
majority of descriptive botanists. 

We proceed with our remarks upon this interesting volume, 
taking up certain points as they strike attention in turning 
the pages, but passing over many others of equal or superior 
importance. 

There is a short chapter upon enigmatical descriptions or 
botanical riddles, and how they come about. The author has 
taken the pains to collect and tabulate the " species dubise " 
of the last four volumes of the "Prodromus," to see who 
is accountable for them, taking into account only botanical 
authors no longer living, and excluding those who have con- 
tributed no more than three. So good a botanist as Blume 
heads the list, one so indifferent as Siebold is accountable for 
the fewest ; so not much comes from such a tabulation. The 
practical point is that Blume, as well as Miquel and Kunth, 



292 REVIEWS. 

who stand high on the list, have fallen much into the habit of 
founding species on fragmentary or quite insufficient herba- 
rium specimens ; instead of passing them over without men- 
tion, or at least without naming them. One is apt to suppose 
that a description of an incomplete specimen, say without flow- 
ers, may be readily eked out later by another hand supplied 
with the missing parts. This, as De Candolle says, is a mis- 
take. The succeeding botanist is hindered more than he is 
helped by such work. And it is the same with species founded 
on figures, such, for instance, as those of Mocino and Sesse, 
upon which the elder De Candolle established species and 
some genera in the earlier volumes of the " Prodromus." As 
to " genera dubia vel non satis nota," very few can be laid 
at the door of first-class botanists. In the list of names of 
deceased botanists which are notable for their absence, the 
name of Torrey is inserted between that of the elder De Can- 
dolle and that of the elder Hooker. 

In the chapter on the description of groups superior to 
species, the author enumerates and sketches the character of 
the six " Genera Plantarum " which have appeared within the 
180 years of modern botany ; the immortal works of Tournefort 
(" Institutiones," 1700), Limmeus (the first edition of whose 
" Genera " was published in 1737), A. L. Jussieu (1789), 
Endlicher (183G-1840), Meisner (1836-1843, which is much 
less known), and finally of Bentham and Hooker, which began 
in 1862, and is now three quarters finished. Tournefort fixed 
the rank and character of genera, Linnaeus tersely and clearly 
defined them, Jussieu arranged them under natural orders, 
defining these, Endlicher and Lindley developed the hierar- 
chy of groups superior to orders, also the tribes inferior to 
them, and the latter is deservedly praised for his sagacity in 
discerning affinities, the former for the perfection of his style ; 
and to Bentham and Hooker is justly awarded the crowning 
merit of having, far beyond their predecessors in this century, 
verified or developed the characters of the genera by a wide 
and direct study of the herbarium materials. 

Floras, or descriptions of natural groups, not in their 
entirety but so far only as represented in a particular country 



DE CAXDOLLE'S PHYTOGRAPHY. 293 

or region, are discussed in chapter X. In Floras, as in more 
general works, abridged descriptions or diagnoses suffice, 
indeed are preferable in all cases where the region has 
been pretty well explored, and where the materials can be 
thoroughly elaborated. Formerly all considerable Floras were 
written in Latin, at least the characters. So they would con- 
tinue to be if the convenience of botanists and the advance- 
ment of science only were to be considered. But Floras are 
used by many to whom even Linnsean Latin would be a 
stumbling-block. Fortunately the difference between good 
botanical English and botanical Latin is not wide, and will 
not seriously trouble a French or German ' botanist. The 
converse hardly holds. The greatest Flora written in English, 
we might say the best great Flora in any language which has 
ever been produced and completed, is Bentham's ' ; Flora 
Australiensis," in seven octavo volumes. Touching upon 
works of special illustration, the " Botanical Magazine " is 
justly singled out for praise, for sustained botanical correctness 
under difficulties, and for its great influence upon the science. 

De Candolle insists much on the importance of describing 
and well classifying the varieties of a species, and of distin- 
guishing them as much as possible into grades, such as sub- 
species or races, varieties, sub varieties, etc. We suggest that 
this can be done with great advantage only when the forms 
are comparatively definite, or have been described as species. 
We think that only the more salient and definite varieties 
should be distinguished by names ; otherwise the names and 
the groups will be limitless. 

In the eleventh chapter, on partial descriptions of groups 
from the point of view of organography (a term which our 
author prefers to morphology), of physiology, botanical geog- 
raphy, etc., our author has some pertinent remarks upon the 
helps which all such studies are offering to phytography, 
which will gradually extend its domain over them ; and upon 
the obvious advantage and great need of having results of 
histological researches expressed descriptively, under some- 
thing like a common terminology, and with due regard to 
rules which have governed the more matured branches of 



294 REVIEWS. 

botany, — rules and practices which eliminate a deal of ver- 
biage, facilitate comparison of views, and ensure mutual in- 
telligibility. Of botanical descriptions for the purposes of 
systematic botany, it could be said that whatever is not clear 
is not botany. May such clearness be hoped for in the future 
of histological botany ? 

Chapter XII. treats of the unavoidable mixture of artificial 
with natural grouping. Truly natural groups are often arti- 
ficially defined, that is, are indicated by single characters ; 
or truly artificial characters are used for the sake of con- 
venience in the division of natural groups. Of the latter 
sort are the divisions Polypetalce, Gamopetalce, and Apetalce 
in Dicotyledons ; also those founded on the mode of curva- 
ture of the embryo in Cruciferce, introduced by Brown, who 
cautiously used them for genera, but raised to the rank of 
primary or subordinal characters by De Candolle. Hypogyny, 
perigyny, and epigyny are in the same category, and probably 
no one was more sensible of it than Jussieu himself, whose 
point and forte was the constitution of orders, not their col- 
location under these artificial heads. De Candolle suggests 
that, while to the more natural divisions are appropriated the 
terms of Class, Cohort, Orders, Tribes, Genera, and Sections ; 
such names as Division, Subdivision, Series, etc., might be 
restricted to artificial divisions, and that these should take 
adjective names not of generic origin, such as Liguliflorce, 
Polypetalce, and the like. 

Chapter XIII. relates to difficulties in phytography which 
have grown out of various methods or absence of method in 
the nomenclature of organs, and from the want of considera- 
tion of the law of priority in such matters. The result of 
which in some departments, such as histological morphology, 
is a state of anarchy not unlike that which prevailed in the 
names of groups before the days of Tournefort and Linnaeus. 
"We may hope that order and lucidity will some day dawn 
upon this chaos and a common language replace this confu- 
sion of tongues. Meanwhile De Candolle offers certain coun- 
sels, the utility of which, he says, is not doubtful nor the ap- 
plication very difficult. 



BE CANBOLLE'S PHYTOGRAPHY. 295 

(1) Hold fast to common and universally known names, 
whether in Latin or in modern languages. Radix, caidis, 
folium, flos, etc., with their vernacular equivalents, are not 
to give place to new-fangled substitutes. This, he thinks, will 
rid us of " such useless terms as caulome, phyllome, etc." 
Now these terms, along with trichome, seem to us legitimate 
and useful, as succinct expressions of a morphological idea ; 
they are annoying only when pedantically ridden as hobbies 
over ground on which they are not wanted. 

(2) Do not entertain the idea that a change in the mode 
of considering or defining an organ requires a change of 
name. Although Linnaeus did take the leaf -blade for the 
leaf, and define it accordingly, that did not much hinder the 
coming in of a truer view, involving merely a change of the 
definition. But one may intimate that De Candolle here 
comes into conflict with another rule he insists on, namely, 
that terms should have unmistakably one meaning. When 
we say — as we ever shall — that leaves are ovate, we speak 
according to the Linnaean definition ; when we say that their 
insertion is alternate, we use the word in a more comprehen- 
sive sense ; when we have occasion to declare that cotyledons, 
bracts, petals, etc., are leaves, we use the word in the most 
comprehensive sense. All this involves considerable ambi- 
guity ; and the endeavor to keep the new wine in the old 
bottles causes no little strain. It is borne because it has 
been applied gradually. If Linnaeus had started with, or 
even reached our ideas, we should happily have had a nomen- 
clature to match. Now we must be content, for descriptive 
purposes, to employ some words both in a restricted and in 
a comprehensive sense, and let the context fix the sense, just 
as it must in ordinary language. Technical precision is only 
a matter of degree. But it is clear that the excellent rule 
here laid down need not forbid the introduction of terms to 
express our conceptions, such as rhizome, caidome, trichome, 
and the like. Yet these are ill-chosen terms, except the last. 
In particular, rhizoma has long ago been appropriated for 
something which is not of root nature, but the contrary. 

(3) The third counsel is to change the name of an organ, 



206 REVIEWS. 

as we do that of a genus or species, only when it is positively 
contrary to the truth, or when it has been pre-occupied. 

(4) Avoid giving special names for rare or ill-definable 
cases of structure. An epithet or short periphrasis is vastly 
preferable to a new and strange term, which will be seldom 
used and may be hardly understood. De Candolle truly re- 
marks that after a great multiplication of terms and distinc- 
tions generally comes some good generalization, which does 
away with a crowd,of particular names ; that what has hap- 
pened in carpology is likely to occur for microscopic organs ; 
and he adds : " Nous assistons au ' feu d'artifice ' d'une 
trentaine de noms de ces etats des cellules " [in our verna- 
cular, we have seen them "go up"] ; "il en restera seule- 
ment quelques-uns generaux ou frequents, qui seront toujours 
necessaires." 

(5) Between two or more names choose, not the most 
agreeable, or even the most significant, but the one best 
known and most widely recognized. 

(6) Between names equally known and used, adopt the 
oldest. Which are the older names is not difficult to know 
in the case of common organs, but is very much so in modern 
histology. 

(7) In this matter of priority or of usage, consider only 
names taken from [or in conformity with] Latin or Greek. 
As in systematic botany, scientific and not vulgar names are 
to be accounted in this regard. Those who like spaltoffming 
for stoma or stomate, and scheitehelle, must needs follow 
their own fashion ; but the genius of our own and the French 
language resists their importation, while it adopts or adapts 
with ease technical terms from classical sources. 

(8) Not to admit names contrary to these rules. 
Chapter XIV. surveys some difficulties in phytography 

which arise from the variant, changed, or contradictory use 
of certain botanical terms, and from the employment of 
vernacular terms which cannot be latinized. The latter has 
just been referred to incidentally. Even the French describe 
the dehiscence of a certain kind of capsule as " en boite a 
savonette." In English we do not attempt to say "in soap- 



BE CANDOLLE'S PHYTOGRAPHY. 297 

box fashion," and should not be understood if we did, but we 
adopt the Linnaean Latin " circumcissile." In general, De 
Candolle concludes that a vernacular term, whether the name 
of an organ or of a botanical group, which will not enter into 
a Latin text by a modification of its termination, is not scien- 
tific, and may give place to one which is. 

A few terms are mentioned which have been more or less 
changed in meaning since the time of Linnaeus ; such as 
lanceolate, which has gradually varied more or less, and for 
a part of the change the present writer is held to account ; 
also glaucus, which classically means sea-green in hue, but 
which has been generally used in botany to designate some- 
times a certain whitishness, and sometimes a whitishness 
caused by a minute Avaxy exudation in the form of a powder : 
the latter is the same as pruinosus. Others may be as sur- 
prised as we were to learn that neither glaucus nor pruinosus 
are Linnaean terms. 

Among the terms used ambiguously, it is surprising that 
De Candolle does not refer to pint ilium, first introduced into 
botany by Tournefort, and used in the sense of the modern 
term gynoeciinn, therefore only one to a flower ; modified by 
Ludwig to denote a female member of the flower (having 
ovary, stigma, and commonly a style), of which there may be 
several or many in a flower ; and adopted in the latter sense 
by Linnaeus, yet generally with a use that avoids contradicting 
the sense of Tournefort. Mirbel, Moquin-Tandon, and St. 
Hilaire among the French, have openly departed from Tourne- 
fort's use, and speak freely of pistils in the plural. Brown 
and De Candolle have used the word in the manner of Ludwig 
and Linnaeus when they have used it at all, but have generally 
evaded its use ; other botanists, especially British, have gone 
back to the Tournefortian sense of gynoeciurn. The present 
writer has a note on the subject in the new edition of his 
" Structural Botany " (1879 and 1880), p. 16G. 

Sinistrorse and dextrorse in the direction of ascent of 
climbing stems or the overlapping of parts in a bud, etc. — 
De Candolle had formerly insisted upon the desirability of 
following what he takes to be the authority and practice of 



298 REVIEWS. 

Linnaeus in the use of these terms ; and lie here returns to 
the subject, reinforcing his former arguments. It is most 
desirable that these terms should not continue to be employed 
in contradictory senses, one party calling that sinistrorse 
which the other calls dextrorse ; it is also fittino; that the 
principle of priority should prevail and that the authority 
of Linnaeus should be respected. Let us, therefore, in the 
first place give an abstract of the points which De Candolle 
here makes. 

But first, we take it for granted that a stem or such organ, 
having no front or back, can have no right or left of its own : 
so when we say that it twines to the left or right, we can 
mean nothing else than the right or left of the observer. The 
contradiction comes from the different position which the 
observer is conceived to occupy. De Candolle supposes the 
observer to be placed within the coil or ascending helix, and 
that this is the more natural position. The other party sup- 
poses the observer to face the object from without ; and from 
this position the Hop twines to the left, i. e., turns in ascend- 
ing from the observer's right to his left, while the Convol- 
vulus turns from his left to his light ; the first is sinistrorse, 
the second dextrorse ; while to De Candolle, standing within 
the coil, the first is dextrorse, the second sinistrorse. Now, 
says De Candolle, Linnaeus in the first edition (1751) of the 
" Philosophia Botanica," § 1G3, page 103, says : " Sinistror- 
sum hoc est quod respicit sinistrum, si ponas te ipsum in 
centro constitutnm, meridiem adspicere ; dextrum itaque con- 
trarium." 

De Candolle remarks that the phrase "meridiem adspi- 
cere" is of no account [but it indicates a certain confusion 
in Linnaeus's mind], for it matters not in what direction you 
look. He adds — what we had all overlooked — that in the 
errata, on p. 3G0, Linnaeus corrected the word sinistrum into 
dextram. But, inasmuch as two editions of the " Philosophia 
Botanica " were printed at Vienna in Linnaeus's lifetime, and 
this correction was not introduced into them, he concludes 
that the correction was cancelled by the author of it. And he 
notes that the expression " sinistrorsum hoc est quod respicit 



DE CANDOLLE' S PI1YT0GRAPHY. 299 

dextram " is a most awkward one for denoting the right-about 
change which the erratum had in view. Nevertheless the 
correction was so made in the edition of the "Philosophia 
Botanica " by Gleditsch in 1T80, two years after the death of 
Linnaeus, also in that of "NVilldenow, published ten years later. 
But De Candolle the elder, in the " Flore Francaise," and in 
all his writings, followed the original text, as also has the 
present De Candolle, who cites as maintaining the same view, 
Braun (who for a time gave way to the opposite), Bischoff, 
Mohl, Palm, Dutrochet, Xaegeli, and even Darwin. But we 
should say that Darwin, noting the conflict of views, had care- 
fully evaded both, using instead the expression " with the sun, 
and against the sun" ; yet sometimes saying " from left to 
right," as equivalent to "against the sun" (as on p. 34), 
showing that he took the external position to be the natural 
one. 

Among those who have used the terms sinistrorse and dex- 
trorse and defined them in the way which supposes the ob- 
server to stand outside of the helix, are Aug. St. Hilaire. 
Duchartre, Bentham and Hooker, Eichler ; and the present 
writer may be added, although our author appears not to be 
aware of it. While trusting that the younger botanists will 
follow the example of Linnaeus and the majority of authors, 
De Candolle recommends that those who depart from it, and 
even those who adopt it, shall state their point of view by 
some convenient abbreviation, such as extus vis. or intus 
vis. ; and thus lessen the danger of a misunderstanding. 
This is indeed essential. 

De Candolle remarks that he can discover no reason for the 
ah extra point of view except a tacit but perhaps nowhere 
expressed assumption that it requires some effort to suppose 
one's self in the centre of a helix or spire. He thinks a mod- 
erate effort will accomplish this. The reply may be that, in 
the case of a stem climbing a hop-pole, or of the scales imbri- 
cated on the axis of a pine-cone, or of a flower-bud on the 
stage of a dissecting microscope, the contemplation of the 
object from without calls for no effort at all. So natural does 
this extraneous position appear to be that we found ourselves 



300 REVIEWS. 

describing these objects from that point of view without think- 
ing of any other, — so natural, as we shall see, that Linnaeus 
fell into it himself, and there remained. Yet, that the oppos- 
ing view has also its fitness is obvious from the fact that the 
physicists and mathematicians are divided in usage, no less 
than the naturalists. 

In the actual state of the case, the question which view 
ought to prevail in botany must be determined therefore on 
a balance of considerations : 1, Priority and authority, such 
as that of Linnaeus ; 2, Naturalness ; 3, Preponderant actual 
usage. We had maintained in this Journal (for March and 
for May, 1877) and in " Structural Botany " (6th ed., note 
on pp. 51, 52) that the externe visum view has decidedly the 
best case on the second ground, and, except in botany, on the 
third also. And now that De Candolle has drawn our atten- 
tion to the matter, we are going to claim the remaining ground 
likewise, and to contend that the contrary usage in botany 
came in from non-attention to the teaching and practice of 
Linnaeus himself. 

On p. 39 of Linnaeus's only own edition of the " Philosophia 
Botanica" he defines and illustrates the directions of twin- 
ing thus : " Sinistrorsum, secundum solem vulgo : Humulus, 
Helxine, Lonicera, Tamus. Dcxtrorsum, contra motum solis 
vulgi ; Convolvulus, Basella, Phaseolus, Cynanche, Euphor- 
bia, Eupatorium." 

Nothing is said about the position of the observer. But 
in every one of the examples of sinistrorse (Helxine being 
Polygonum convolvulus'), the stem winds around the support 
passing from right to left of the observer confronting the coil; 
and in every one of the dextrorse examples (Eupatorium 
being Mikania) it winds in the opposite direction. That is, 
dextrorse and sinistrorse are used in the externe visum sense. 
On p. 103 the same is repeated, except that reference to the 
sun's apparent course is omitted and additional examples are 
added, most (but not all) of them accordant with the preced- 
ing. So far, it would seem that Wichura was not mistaken 
in his statement that De Candolle had followed a different 
method from that of Linnaeus. And this appears to be the 
whole case as respects direction of twining. 



DE CANDOLLE'S PHYTOGRAPHY. 301 

But on the same page, to " Corolla sinistrorsum " is ap- 
pended the foot-note which has made so much trouble, namely, 
" Sinistrorsum hoc est, quod respicit sinistrum, si ponas Te 
ipsum in centro constitutum, meridiem adspicere ; Dextrorsum 
itaque contrarium." That is to say, in defining the direction 
of overlapping of the parts of a perianth, Linnaeus took the 
open flower instead of the bud, and proposed to look down 
upon it from above or within. Now it may well be that Lin- 
naeus subsequently perceived the contradiction between his 
terminology for overlapping and that for twining ; and that 
his brief erratum, on p. 310, " pro sinistrum lege dextram," 
was intended to bring the former into congruity with the lat- 
ter, which it does, but in an awkward way. Perhaps he saw 
the incompatibility of the cited examples ; in fact, about as 
many of them accord with the outside as with the inside point 
of view. Any way, the erratum is his own ; it seems unlikely 
that he authorized its omission from the Vienna editions ; and 
Gleditsch and Willdenow should not be blamed for heeding 
his behest in their editions. For, so far as it goes, it tends to 
render their author consistent with himself. If Linnaeus had 
revised the page himself, he would have left out the " meri- 
diem adspicere," which has nothing to do with the matter, 
and doubtless he would have completed his assimilation of 
the direction of petal-obliquity or overlapping with that of 
stem-windins: ; and so the whole confusion from which we 
are endeavoring to escape would have been avoided. 

In adopting the external point of view — now fortified by 
original authority — it is well to note that we shall be in 
accord with the modern physicists and mathematicians, and 
also with common people. The ordinary screw, on which the 
thread winds from left to right of the confronting observer, 
and which is driven home by the semi-rotation of the hand 
and fore-arm from left to right, is everywhere known as the 
right-handed screw ; and this, with the corkscrew, is taken as 
the norm and exponent of right-handed rotation by Clerk- 
Maxwell (" Treatise on Electricity and Magnetism," i. 23), 
and by Sir Wm. Thomson. 

The analogies which have been adduced in favor of the 



802 REVIEWS. 

inside position are mostly drawn from objects which have 
a right and left of their own ; a building, for instance, has 
a riarht and left side or wing because it has a front and a rear. 
The right side of an assembly presided over by an officer who 
faces the members is quite arbitrarily, but naturally, taken to 
be that on the right of the chairman. But the right-hand 
figures on a drawing or engraved plate are taken to be those 
on the right hand of the observer, notwithstanding that the 
plate, having face and back, has a right and left of its own. 

Chapter XV. refers to certain difficulties which grow out 
of ambiguous terms of ordinary language ; for example, the 
various meanings of the word (fin) end or purpose, and the 
ambiguities in the use of the terms nature, natural, super- 
natural (which lead off into philosophy, but are here treated 
rather in reference to style of exposition) ; also the change 
which has occurred in the scope of the word history in natural 
science. 

Chapter XVI. is an interesting and pertinent one, upon the 
manner in which facts observed under the microscope are 
described, and on the great saving of space and advantage 
in clearness which would be gained by the adoption, for all 
matters perfectly capable of it, of the Linnsean descriptive 
style, and of Linnsean Latin. Extracts from the German of 
Schacht, the French of Payer, and the Italian of Gasparrini 
are given, and by their side a rendering in descriptive Latin ; 
and the words and letters are counted. The German specimen 
so treated is diminished to considerably less than half the 
number of words and a little less than half the number of 
letters. The French simmers down to one third the number 
of Latin words and less than half the number of letters ; and 
in the French of descriptive botany to less than one half. 
The Italian extract of 51 words and 256 letters is expressed 
in Latin of the Linnsean form by 21 words and 127 letters. 

Style in botanical works is discussed in chapter XVIII., 
which all young botanists should study, especially the portion 
which treats of the admirable style of Linnaeus. In speaking 
of botanical style in the modern languages, the author notices 
the great advantage which the languages of Latin stock have 



BE CANDOLLE'S PHYTOGRAPHY. 303 

inherited, and which the English-writing botanists have ac- 
quired, of ready and free use of Latin and latinized technical 
words by direct transference. Botanical French, English, 
and Italian, are contrasted with the German in this respect. 
Noting that the German of conversation inclines to be clear 
and sententious, while in botanical writings the words lengthen 
more and more and the sentences become badly involved, our 
author remarks that recently having read a couple of pages of 
Vegetable Anatomy, and feeling his brain somewhat fatigued 
with the frequency of such words as Sclerenchymfiisergrup- 
pen, Gefiissbundentwicklung, and Entwicklungseigenthuni- 
lichkeit, he asked himself if that was good German style. 
He then recollected that Goethe, one of the very greatest of 
German literary writers, was also a profound naturalist. He 
opened his "Metamorphose der Pnanzen," read a page or so, 
and experienced a relief which he likens to that felt by a sea- 
tossed ocean voyager when the vessel suddenly glides into 
a quiet harbor. 

Chapter XIX. discusses the propositions to employ letters 
and figures, chosen arbitrarily or otherwise, to represent spe- 
cific and generic characters, — repulsive contrivances, to which 
our author lends no countenance. 

Chapter XX. treats questions of orthography, abbreviations 
and signs, pagination, typography ; the twenty-first chapter, 
of titles and indexes ; both full of interesting details upon 
which we cannot touch, although we are longing to put in 
our oar. 

Chapter XXII. animadverts upon the tendency of certain 
modern cryptogamists to set all botanical rules at naught. 
The next gives advice about articles in journals, dissertations, 
and the like ; the next treats of translations ; another, of fig- 
ures, and has many noteworthy remarks ; chapter XXVL, of 
auxiliary and bibliographical works ; and chapter XXVII. is 
a chronological table of the progress of phytography, begin- 
ning with a Chinese encyclopedia 1000 years before Christ, 
and ending with Sachs's " Lehrbuch," 1868-1877. Botanical 
students will find it very interesting and instructive. 

The remaining chapter begins the second part of the vol- 



304 REVIEWS. 

ume, " Prcuves des Descriptions ; " which is principally de- 
voted to herbaria, their history, formation, and management ; 
— a most important chapter, the analysis of which would 
form an article by itself. Last in order and not least in im- 
portance, a full enumeration is given of botanical collectors 
and authors who have formed herbaria, with an indication of 
the place where their herbaria or collections are preserved. 



DARWIN'S POWER OF MOVEMENT IN PLANTS. 

First let us congratulate the scientific community, no less 
than the author, that Mr. Darwin's experimental researches 
are seconded, and are we hope long to be continued, by the 
son whose name appears upon the title-page, and whose in- 
dependent papers already published approve his worthiness 
for that honor. This volume 1 is from beginning 1 to end a 
record of a series of researches and of the inferences which 
they directly warrant. Naturally it will not fascinate the 
general reader after the manner of "The Origin of Species" 
and some of the volumes which succeeded that epoch-marking 
production; nor has it the fresh charms of the treatises "On 
the Movements and Habits of Climbing Plants," and of 
" Insectivorous Plants," of which it is the proper continuation 
and supplement. 

The organs of plants take certain determinate positions 
and execute certain movements, some of them universal or 
general, some of them special, some of them very striking and 
seemingly strange, most — but not quite all of them — evi- 
dently advantageous to the plant or essential to its wellbeing. 
Roots point toward the earth ; stems point away from it ; 
young stems bend toward the light, and the upper face of 
leaves is presented to it. Stems that twine " circumnutate " 
(a capital term), i. e., bend successively to all points of the 

1 The Power of Movement in Plants. By Charles Darwin, assisted by 
Francis Darwin. With illustrations. London and New York, 1880. 
(American Journal of Science and Arts, 3 ser., xxi. 245.) 



POWER OF MOVEMENT IN PLANTS. 805 

compass, arid this wholly irrespective of external influences ; 
and the twining around a support is the direct consequence 
of the circumnutation. Most tendrils freely circurnnutate, 
and thereby are enabled to reach the object which they grasp. 
Most tendrils (and in certain cases some other parts) are 
very obviously sensitive to external contact or irritation, to 
which they respond by movement and change of form, and 
thus they grasp or do other advantageous acts. Some move- 
ments, especially of leaves, occur with regularity upon the 
access of light, others with its withdrawal ; a few, such as the 
small leaflets of Desmodium gyrans, proceed irrespective of 
night and day. The specification need not be extended. The 
general facts in all their great variety are familiar to scientific 
readers. The inquiry of this volume is as to their ground 
and origin, or, as in this connection we should rather say, 
their development and history. For instance, circumnutation 
gives rise to twining and gives efficiency to other ways of 
climbing. But Darwin is bound to suspect, and even to show, 
that circumnutation is not a special endowment of the stems 
and tendrils of climbing plants, but rather a more developed 
manifestation of a general faculty. And the same is to be 
said of the movements of tendrils and leaves, or their appen- 
dages, whether automatic or in response to external irritation 
or stimulus. All this is what the experimental researches 
detailed in this volume undertake to ascertain and have satis- 
factorily made out. 

An abstract of the volume might be somewhat tedious, and 
is certainly unnecessary for biological readers, who are sure 
to possess and study it. But the gist is readily to be gathered, 
without running through the iterated details or scanning 
many of the illustrative and curious figui*es which record the 
movements under investigation, by the simple perusal of the 
introduction and of the concluding chapter, in which the mat- 
ter of the volume is summed up. 

The sum and substance of the case is, that all these powers 
and faculties are manifested in the seedling immediately upon 
germination, and most of them are then remarkably exem- 
plified. The caulicle or initial portion of stem below the 



80G REVIEWS. 

cotyledons (with the elongation and protrusion of which the 
germination of dicotyledonous seeds usually begins) circum- 
nutates as soon as it conies out into the open air, and even 
earlier: this is the earliest manifestation of an automatic 
movement which is shared by all the succeeding portions of 
stem developed from it, in the early life of most plants, 
whether climbers or not. In the latter, and especially in 
twining plants, we see this general faculty at its maximum 
and in beneficial exercise. More remarkable and novel it is 
to learn that the initial root, growing from the lower end of 
the caulicle (not inaptly called by Darwin the hypocotyl'), also 
shares in this faculty of circumnutation. As it penetrates 
the soil in its downward course, it cannot largely manifest 
this faculty, and indeed its power of circumnutation is always 
small ; " but the circumnutating movement will facilitate 
the tip entering any lateral or oblique fissure in the earth or 
a burrow made by an earth-worm or larva ; and it is certain 
that roots often run down the old burrows of worms. The 
tip, however, in endeavoi-ing to circumnutate will (succes- 
sively) press against the earth on all sides, and this can hardly 
fail to be of the highest importance to the plant " (being sup- 
plemented by another faculty, that of sensitiveness at the tip 
presently to be mentioned) ; for " when the tip encounters a 
stone or other obstacle in the ground, or even earth on one 
side more compact than on the other, the root will bend away 
as much as it can from the obstacle or the more resisting 
earth, and will thus follow with unerring skill the line of least 
resistance." Then, beside the almost universal heliotropic 
movement, by which each leaf or leaflet presents its superior 
surface to the direction of the greater light, Mr. Darwin 
shows that these organs also circumnutate, beginning even 
with the cotyledons or seed-leaves ; although their sweeps gen- 
erally form so narrow an ellipse that they move up and down 
in nearly the same vertical plane, a movement describing a 
circle being converted into one up and down. 

These circu'mnutatory movements are of the most funda- 
mental and therefore mysterious character. Although most 
commonly connected with growth, they are at bottom inde- 



POWER OF MOVEMENT IN PLANTS. £07 

pendent of it. This — contrary to some German physiolo- 
gists — we must conclude from both DeVries' and Darwin's 
investigations. They are produced by the changing tumes- 
cence of the cells on different sides of a stem or footstalk, 
which may or may not be fixed by some consequent growth 
or solidification. This Mr. Darwin, we presume rightly, con- 
cludes to be the faculty or susceptibility upon which heliotro- 
pism, geotropism, and the like (not to speak of apheliotropism, 
apogeotropism, parahdiotropism, diaheliotropism, hyponasty, 
nyctotropism, and other terms which the incautious student 
may take to be powers instead of abbreviated expressions) — 
in other words, upon which the solar rays and some occult in- 
fluence of the earth — act, modifying the sweeps or converting 
them into forth and back or other special movements. Among 
these, that which has been termed the sleep of leaves, better 
and briefly designated by the word nyctotropism, is thor- 
oughly investigated in this volume, and is shown to be far 
more general than has been supposed ; and the conclusion is 
that the end subserved is a needful protection of the surfaces, 
mainly of the superior surface against cold from nocturnal 
radiation. A priori, looking at the structure of the leaf, one 
would have thought that the under surface had the greater 
need of such protection. 

Not only are all these movements incipient in the seedling, 
but some of them are manifested more rapidly and extensively 
than in most mature plants. This should needs be, since, as 
Mr. Darwin states it, " Seedlings ai*e subjected to a severe 
struggle for life, and it appears to be highly important to 
them that they should adapt themselves as quickly and as 
perfectly as possible to their conditions." Very properly, 
therefore, no small part of this volume is devoted to the seed- 
ling and to the behavior of its several parts. The most novel 
and unexpected results relate to the young root. Judging 
from its simplicity and from the medium in which it is de- 
veloped, one would not look there for the endowment which 
Mr. Darwin finds in it. But this root-tip and the vegetable 
cells which compose it conspire to teach us that the most sim- 
ple structures may be wonderfully gifted. The tiny root ex- 



308 he views. 

hibits three kinds of movement : first that of circumnutation, 
in which, endeavoring to bend in all directions, its tip " will 
press on all sides, and thus be able to discriminate between 
the harder and softer adjoining surfaces, . . . and to bend 
from the harder soil and follow the lines of the least resist- 
ance," so modifying advantageously its course from that to 
which geotropism constantly tends to give it. Moreover, the 
growing end of the root is sensitive to contact, and in a com- 
plex manner. If pressed above the tip, it bends there toward 
or around the impinging body, much as the end of a tendril 
bends around a support : thus it may follow, as roots do, along 
the unequal surface of a solid body. But, thirdly, if the tip 
itself be locally pressed, it exhibits different and more sur- 
prising sensitiveness, for it transmits an influence to an upper 
adjoining part, causing it to bend away from the affected side. 
This sensitiveness to contact is confined to a little more than 
one millimeter of the tip ; the part which bends is 6 or 7 or 
even 12 millimeters above. So, when the sensitive tip in its 
downward growth strikes obliquely upon a stone or other ob- 
stacle, the part above at this distance, to which some influence 
must be transmitted, bends and carries the point away from 
the obstacle. Yet later, when a new portion of the side im- 
pinges upon the stone or other body, it will bend at that part 
toward instead of away from it, and so follow along its sur- 
face. It is the tip, likewise, that can discern that the air is 
moister on one side than on the other, and which thence " trans- 
mits an influence to an upper adjoining part, which bends 
toward the source of moisture." It is the tip only which is 
sensitive to gravitation. Well may Mr. Darwin affirm that 
there is no structure in plants more wonderful, as far as its 
functions are concerned, than the tip of the radicle. Also, 
that " it is impossible not to be struck with the resemblance 
between the foregoing movements of plants and many of the 
actions performed unconsciously by the lower animals." " But 
the most striking resemblance is the localization of their sen- 
sitiveness and the transmission of an influence to an excited 
part which consequently moves. Yet plants do not of course 
possess nerves or a central nervous system ; and we may infer 



POWER OF MOVEMENT IN PLANTS. 809 

that with animals such structures serve only for the more per- 
fect transmission of impressions and for the more complete 
intercommunication of the several parts." The closing sen- 
tence of the book may be appended to this. " It is hardly an 
exaggeration to say that the tip of the radicle, thus endowed 
and having the power of directing the movements of the ad- 
joining parts, acts like the brain of one of the lower animals." 

The movements " excited by light and gravitation," as well 
as the nyctotropic or sleep-movements so called, are (as we 
have already stated ) all referred by Mr. Darwin to modified 
circumnutation, "which is omnipresent while growth lasts, 
and after growth has ceased whenever pulvini are present," 
as in several classes of leaves. As respects the relation of 
external agents to the movements, note Mr. Darwin's re- 
mark : " When we speak of modified circumnutation we 
mean that light, or the alternations of light and darkness, 
gravitation, slight pressure or other irritants and certain innate 
or constitutional states of the plant, do not directly cause 
the movement ; they merely lead to a temporary increase or 
diminution of those spontaneous changes in the turgescence 
of the cells which are already in progress." 

Certain parts of plants turn or grow earthward. When 
this is attributed to gravitation, as it commonly is, the physi- 
cists have opportunity to complain of the misuse of a term. 
Although Mr. Darwin, like other writers, speaks of the in- 
fluence of light and of gravitation in the same breath, without 
discrimination, we note with satisfaction his disagreement with 
those who " look at the bending of a radicle towards the 
centre of the earth as the direct result of gravitation," and 
note especially the closing dictum. " Gravity does not ap- 
pear to act in a more direct manner on a radicle than it does 
on any lowly organized animal, which moves away when it 
feels some weight or pressure." Why, we would ask, need 
the word gravity or gravitation be used at all in this con- 
nection ? 

The introduction to this volume contains a short article 
upon the terminology which is adopted in it, chiefly as re- 
gards such words as epinasty and hyponasty, geotropism and 



310 REVIEWS. 

related terms, which is most convenient to employ, and also 
the names of these several parts of the embryo and seedling. 
This is, we believe, almost the first English book in which the 
axial part of the dicotyledonous embryo below the cotyledons 
(the radicle of the systematic botanist even of the present 
day) is distinctly recognized as hypocotyledonous or initial 
stem, although on the continent and in America this has long 
been taught and accepted. None the less so although the 
term radicle has been retained for it (until recently by the 
present writer, at least), in order not to break with the ter- 
minology of systematic works. Mr. Darwin, in this volume, 
shortens the expression of " hypocotyledonous stem " into the 
term " hypocotyl," — a fairly good English term, certainly 
better than the French tigelle. The objection to them both is 
that the words will not take the substantive Latin form, as all 
such terms should. Wherefore the better name — an old one 
which we have reverted to in the last edition of the Botanical 
Text-book (Structural Botany) — is caulicle or cauliculus. 
The initial root, which grows from the lower end of the 
caulicle (or " hypocotyl "), Mr. Darwin calls the radicle, fol- 
lowing in this the ordinary English usage, except in very 
definitely distinguishing it from the cauline part above it. 
Being simply root, we have preferred uniformly to call it so, 
thus avoiding a word which the systematists have all along 
applied to the caulicle. Although initial stem and initial 
root are most clearly discriminated in the present volume, yet 
in the accounts of the germination or the ordinary Dicoty- 
ledons, it appears to be implied or stated, either that it is the 
root-part which first projects from the seed-coats and that the 
stem-part begins its development later, or that the axial part 
of the embryo conspicuously preexisting in the seed is root 
and not stem. We take it to be quite otherwise, namely, 
that this axial part in the seed is cauline, and that ordinarily 
it protrudes or makes some growth in length before root-for- 
mation begins. 

A few misprints of names of plants will in no wise mislead 
or trouble any botanist, except possibly in the case of Aphnn- 
graveohns, which on p. 422 and 424, and in the index, is 
printed Apios. 



ORIGIN OF CULTIVATED PLANTS. oil 



DE CANDOLLE'S ORIGIN OF CULTIVATED PLANTS. 

M. Alphonse De Candolle's " Geographie Botanique 
Raisonnee," in two volumes of nearly 700 pages each, was 
published in the year 1855, and has been for several years 
out of print. It is not surprising that the now venerable but 
still well-busied author should decline the labor of preparing 
a new edition, involving, as it would, the re-discussion of cer- 
tain questions under changed points of view, and the colloca- 
tion of a vast amount of widely scattered new materials which 
the last quarter of a century has brought to us. 

Happily, the chapter on the geographical origin of the 
species of plants generally cultivated for food, and for other 
economical uses, could be detached. This, the author has 
sedulously studied anew ; and the present volume 1 is the re- 
sult. As yet we have it only in the original French ; but it 
is said that an Euglish translation is in preparation. So, if 
the work is not already in the hands of botanists and other 
scholars generally, we may expect that it soon will be ; and, 
contenting ourselves with a mere mention of its plan and 
scope, we may proceed to remark, here and there, upon points 
which strike our attention. 2 "We may expect this to be for 

1 Origine des Plantes Cultive'es. Par Alphonse De Candolle. Paris, 
1883. (Bibl. Scientifique Internationale, XLIII.) (American Journal of 
Science and Arts, 3 ser., xxv. 241, 370 ; xxvi. 128 ; with J. Hammond 
Trumbull.) 

2 To avoid repetition, it may be mentioned here that, iu the follow- 
ing annotations, the " Relations of the Voyages " of Columbus are cited 
from Navarrete's "Coleccion de los Viajes," etc. (Madrid, 1858, and 1827- 
37) ; references to Peter Martyr d'Anghiera's first three Decades " De 
Rebus Oceanicis et Novo Orbe " (written before 1517) are to the Cologne 
edition of 1574 ; references to Oviedo's " Historia General y Natural de las 
Indias " — of which the first nineteen books, published in 1535, included a 
revised and enlarged edition of his " Relacio sumariade la Nat. Historia de 
las Indias," printed in 1526 — are to the edition published by the Royal 
Academy of History of Madrid, 1851-55 ; Jean de Lery's " Histoire 
d'un Voyage faict en la terre du Brasil " (in 1557-8) is referred to in his 
revised edition in Latin, " Historia Navigations in Brasilian " (Geneva?, 
15S6) ; Fr. Hernandez, "Nova Plantarum, etc., Historia," in the edition of 



312 REVIEWS. 

many years the standard work upon the subject, and to 
undergo revision in successive editions ; and we are sure that 
the excellent author will welcome eveiy presentation of dis- 
cussion which may chance to throw any new light upon the 
sources or the aboriginal cultivation of certain plants which 
the Old World has drawn from the New. 

The first part of the volume, of only 22 pages, is mainly 
occupied with a consideration of the means employed for the 
determination of the sources whence the various cultivated 
plants have been derived. The botanist enquires where a 
given cultivated plant grows spontaneously, or what was its 
wild original ; and he has to judge, as well as he can, where 
it is truly indigenous or where a reversion from a cultivated 
to a wild condition. This, as respects weeds and the like, is 
a difficult matter, even in a newly settled country like North 
America, much more so in the Old World ; but as respects 
the plants of agriculture, the case is usually simpler. The 
botanists resident in a country are not likely to be far misled 
by the occurrence of wilderings ; but, in the case of travel- 
ers and collectors, perhaps too much has been made, even 
in this volume, of plants only once met with growing spon- 
taneously, and inferred to be indigenous. Plentifulness is 
of no account, else the Century plant and Opuntia would be 
thought indigenous to the Mediterranean region, the Ox-eye 
Daisy to the United States, and certainly the Cardoon to the 
Pampas, where there is now probably more of it than any- 
where in the Old World. Archaeology and palaeontology 
are often helpful, as by the identification of fruits and seeds 
in ancient Egyptian tombs, or of paintings upon their walls, 
or of fragments in ancient bricks ; or the debris of lake- 
dwellings rescued from lacustrine deposits, as in Switzer- 

Rome, 1G51 ; " Rariorum Stirpium Historia " by L'Ecluse (Clusius), in 
the first edition, Antwerp, 1576 ; his " Exotica," including his translations 
of Monardes and Acosta, Antwerp, 1605, with his "Curse Posteriores " 
(posthumous), 1611. — J. H. T. 

II va sans dire — yet should explicitly be said — that all the historical 
and philological lore, which gives this article its value, is contributed 
by my associate. — A. G. 



ORIGIN OF CULTIVATED PLANTS. 313 

land ; and from the tufas of southern France, the kioken- 
moddings of Scandinavia, the mounds of North America, and 
the ancient monuments and tombs of Mexico and Peru. His- 
torical documents are also important for the date of certain 
cultures in particular countries ; and here it is stated that 
the principal cultures have come from three great regions, 
namely: China, southeastern Asia and Egypt, and intertro- 
pical America. De Candolle also remarks that in the Old 
"World agriculture was developed along rivers, in the New, 
upon plateaux, — a fact which he attributes to the primitive 
situation of certain plants worth cultivating. But this is not 
quite obvious. Linguistic learning may be turned to much 
account ; as in tracing a plant toward its home by the name 
which has gone forth with it in all its migrations. Like other 
instruments this must be used with some knowledge and judg- 
ment. Ble de Turquie (maize) did not come from, and 
probably not by way of, Turkey, any more than did the 
animal of that name. Jerusalem Artichoke has naught to 
do with Jerusalem, but came from North America, and is no 
artichoke. Pomme d' Acajou, anglice Mahogany-apple, is 
neither an apple nor a pomaceous plant, nor has it anything 
to do with mahogany. New Zealand Flax came indeed from 
New Zealand, but is not a flax. Among errors from the 
careless transference of names from one plant to another, 
that of Potato, which belongs to the Batatas or Sweet Potato, 
is familiar. Of mistakes which have been made in the trans- 
ference of a popular name from one language to another, 
De Candolle mentions the Arbre de Judee of the French, 
which in English has become Judas-tree. We may add that 
of Bois fidele, of the French West Indians, which, taken up 
by their English successors as Fiddle-wood, has been perpet- 
uated in the generic name Citharexylum. 

The several lines of evidence — botanical, archaeological, 
palaeontological, historical, and linguistic — may be used to 
supplement or correct each other. How they may be brought 
to bear, and how their combination may give satisfactory 
results, is practically shown in Part II, — a study of the species 
as regards their origin, their earliest culture, and the principal 



314 REVIEWS. 

facts of their dispersion, — which makes up the principal hulk 
of the volume, namely, from page 23 to page 350. 

This part is divided into chapters, e. g.. Plants cultivated 
for their subterranean parts, such as roots, bulbs, tubers, etc. 
Those cultivated for their herbage, whether for human food, 
for forage, for fibres, for stimulation, etc. ; but the proper 
medical plants are left wholly out of view, as likewise plants 
cultivated for ornament. So the chapter on plants cultivated 
for their blossoms, or parts connected with these, is brief 
enough, treating as it does only of the Clove, Hop, Safflower, 
and Saffron. For the Rose, Acacia Farncsiana, and all plants 
however largely cultivated for perfume or for essential oils are 
left out of view. So also are the sweet-herbs of the kitchen- 
garden, and all condiments, except Horse-radish. Plants cul- 
tivated for their fruits and seeds occupy the closing chapters. 
Among the latter the Cotton-plant is placed. The arrange- 
ment matters little, and that adopted may be the most con- 
venient. A good index makes ready reference to any topic. 

In the order of the book we come first to Ilelianthus tube- 
rosus, the Topinambour of the French, Jerusalem Artichoke 
of the English ; in the United States the tubers simply called 
artichokes. Almost all we know of the origin and source of 
these esculent tubers has been recovered since the publication 
of De Candolle's earlier work, in 1855. Although the con- 
temporary accounts specified its introduction from Canada, 
and Linnaeus so cites it in the " Hortus Cliffortianus," the sub- 
sequent reference to Brazil was followed without question 
down to De Candolle's " Prodromus " ; and the present author, 
in the work above mentioned, doubted the Canadian as well 
as the Brazilian origin. It now appears that Schlechtendal 
(in Botanische Zeitung, 1858) was the first to recover a part 
of the documentary history. Our own article on the subject 
— to which there is nothing of importance to add — was con- 
tributed to this Journal for May, 1877. 1 Singularly, it has 

1 In it reference was made to Lescarbot's mention of roots "grosses 
comme naveaux . . . ayans an gout retirant aux cardes," etc., and cited 
his " Histoire de la Nouv. France," in the edition of 1G12 (p. 840). In a sub- 
sequent edition (1G18), cited by M. De Candolle, Lescarbot adds that he 



ORIGIN OF CULTIVATED PLANTS. 315 

remained unknown to De Candolle, although it is referred to 
at the close of Decaisne's independent and exhaustive article, 
in the " Flore des Serres," 1881. 

It can now be said that the wild plant to which Helianthus 
tuberosus has been traced is not H. doronicoidcs, Lam., 
although it was confounded with that species in Torrey and 
Gray's " Flora." Lamarck's plant is a sessile-leaved species. 
Decaisne's remark that //. tuberosus is the only species of 
the genus which is at all tuberiferous may be qualified. A 
form of what appears to be H. giganteus, but is not yet very 
well known, grows in Minnesota and the Saskatchewan region, 
has been mentioned by Douglas under the name of " Indian 
Potato " of the Assiniboine tribe, by Bourgeau as " H. sub- 
tuberosus" in herb. Kew, and by Dr. C. C. Parry in Owen's 
Minnesota Report, page 614, under the name of //. tuberosus. 
The scanty tubers which we have seen in dried specimens do 
not compare well with those of //. tuberosus ; and that species 
has never been found wild so far north (that we know of), 
not even in the most southern parts of Canada West. The 
aborigines who cultivated it must have obtained it from the 
valleys of the Ohio and Mississippi and their tributaries, 
where it abounds. 

had brought these roots into France, where they began to be sold under the 
name of Topinaraboux, and that their Indian name was Chiquebi. On this 
last point, Lescarbot was wrong. Chiquebi was an eastern Algonkin name 
for the tubers of Apios tuberosa, the common "groundnuts," — not for 
those of Hdianthus tuberosus. It is easy to see how Lescarbot was misled. 
Father Biard's "Relation de la Nouv. France " was printed in 1616, and 
in it (chap. 22) there is mention of certain " racines, appellees en Sauvage 
Chiquebi," which grow spontaneously under oaks : " elles sont comme des 
truffes, mais meilleures, et croissent sous terre enjilces Vune a Vautre en 
forme de chapelet," etc. Lescarbot doubtless caught the name from Biard, 
and misapplied it. Father Paul Le Jeune (Relation, 1634, chap. 7) men- 
tions these ground-nuts, " une racine que nos Francois appellent des chape- 
lets, pource qu'elle est distingue'e par nceuds eu forme de grains." Les- 
carbot's " Topinamboux " indicates a popular belief, in France, in the 
Brazilian origin of H. tuberosus. The Tupinamba Indians of Brazil — a 
division of the Tupi-Guarani family — had been allies of the French in 
the sixteenth century, and their name was probably well known in France 
through the relations of J. de Lery and other voyagers. Lescarbot 
(Histoire de la Nouvelle France, 1612, p. 178) follows Lery in writing 
the name " Tououpinambaoult." — J. H. T. 



316 REVIEWS. 

Helianihus annuus, L. — the history of which was almost 
equally confused, and which we had identified with a wide- 
spread species of the western United States — is omitted by 
De Candolle, yet might claim a place : for Decaisne, who has 
treated it at length in the paper above cited, informs us that 
a form of it (called " Russian Sunflower ") is cultivated in 
Russia for the oil of its large seeds, and, if we mistake not, for 
fattening poultry. Our Indians also cultivated it for the oil 
of the seeds, which they used for greasing their hair, also for 
eating and other purposes. Champlain noted this (in 1610?), 
and Sagard about a dozen years later. 1 The latter says (His- 
toire du Canada, 1736, page 785) : "lis font estat du tourne- 
sol, qu'ils sement en quantite, en plusieurs endroits a cause de 
l'huyle qu'ils tirentde la graine," etc., piously adding: "Mais 
comment est-ce que ce peuple sauvage a pu trouver l'invention 
de tirer d'une huyle que nous ignorons, sinon a l'ayde de la 
divine Providence." The wild original of this Sunflower must 

1 Cbamplain's earlier record of the cultivation and use of the Sunflower 
is essentially like that of Sagard, and both relate to the same stations, 
namely, the Huron towns near the southeastern point of Georgian Bay. 
This Champlain reached by way of the Ottawa (R. des Prairies) and Lake 
Nipissing. The lamented Decaisne has here introduced some confusion 
into the history, which we hasten to rectify. In his article in the "Flore 
des Serres" (xxiii. p. 108, p. 2 of the pamphlet), he says, " Je trouve 
dans Champlain l'observation suivante (Voyage Nouv. France, re'impress. 
1830, torn. i. p. 110) : " 

" En remontant le St. Laurent et avant l'arriver au Lac Ontario, je 
visitai cinq des principaux villages ferule's de palisades de bois, jusqu'k 
CahiagueV' etc., and so on to the mention of the " grande quantite' de bled 
l'Inde (Mais) qui y vient tres beau, comme aussi des citrouilles. Herbe des 
soleil, dont ils font de l'huile, de la graine de laquelle ils se frottent la 
tete." 

Tbis, the latitude of 44.50° being stated, would refer Cahiague* and the 
Sunflower cultivation to the neighborhood of Ogdensburgh and Prescott, 
far away from the actual place (the Indian town mentioned being the 
Huron name of the mission station of San Jean Baptiste, in what is now 
Simcoe Co.), and it introduces a palpable anachronism, Ontario having 
been an unknown name in Cbamplain's time. In fact, there is nothing 
answering to the early part of this pretended quotation, either in the orig- 
inal of Champlain or in the edition here cited by name and page. The 
excellent Decaisne could never have tampered with the quotation himself. 
He must have taken it at second hand and neglected to verify it. 



ORIGIN OF CULTIVATED PLANTS. 317 

have been obtained by the Canadian Indians from beyond the 
Mississippi, and some degrees farther south. Judging from 
the breadth of the flower-heads soon after its introduction into 
Europe, it must in aboriginal hands have assumed much of 
the abnormal development which distinguishes the cultivated 
Sunflower from its wild original of the western plains. 

Solatium tuberosum, L. — The question of the Potato was 
fully discussed by De Candolle in 1855 ; and the present 
review of it only confirms the now generally admitted conclu- 
sions. These are summed up in the statements, that the plant 
is spontaneous in Chili under a form quite identical with the 
cultivated species, that its aboriginal cultivation had extended 
as far north as New Granada, but apparently no farther ; that 
allied tuberiferous species, which our author regards as dis- 
tinct (though others partly doubt it) are found along the 
Andes and through Mexico, and within the borders of the 
United States ; that when known in Virginia and North Caro- 
lina in the second half of the sixteenth century, it was not 
derived from our Indians ; and that it was carried to Europe 
first by the Spaniards between 1580 and 1585, and afterwards 
by the English. 

Batatas vulgaris. Choisy, Convolvulus Batatas, L., the 
Sweet Potato, is one of a few cultivated plants which have at- 
tained to a very wide distribution over the warmer parts of the 
world in early times ; and it is one which no botanist pre- 
tends to have seen in a truly wild state. The evidence inclines 
to an American origin ; but it had reached the Pacific islands 
in prehistoric times, and was cultivated in China in the second 
or third century of our era. De Candolle states that — 

" Clusius, one of the first to speak of the Batatas, says that 
he had eaten it in the south of Spain, where it was said to 
have come from the New World. He indicates the names of 
' Batatas,' ' Amotes,' ' Ajes.' " 

The testimony of Clusius (L'Ecluse) to the American origin 
of the Sweet Potato, though not of the highest value, might 
be more strongly stated. He visited Spain and Portugal in 
1566. The first edition of his " Historia Eariorum Stirpium " 
was printed in 1576, and contains the description of Batatas, 



318 REVIEWS. 

which M. De Candolle cites from the edition of 1601. He 
gives a figure of the plant, of which, he says, he had observed 
three varieties growing in the south of Spain. He states 
their American origin, not as a doubtful matter or with a 
" Ton pretendait," but as a well-established fact : " Sponte 
nascitur in novo orbe, vicinisque insulis, unde primum in His- 
paniam delata est." "Now," he adds, "it is planted in many 
places near the coast of Andalusia ; but those grown at Mal- 
aga are preferred, and are transported to Cadiz and Seville. 
We sometimes have them fresh in Belgium, but they will not 
germinate here, the country being too cold." As to the name 
— he was as undecided as have been some botanists since his 
time : " the Spaniards call them ' Batatas,' and also ' Ca- 
motes ' or ' Amotes ' ; some also ' Ajes ' ; yet, as they say, they 
differ among themselves, and the root of Batatas may be 
much the sweeter and the more tender." 

This confusion of names dates from the time of Columbus, 
for Clusius was not, by half a century, the first to speak of 
the Batata. (It may be worth noting, in parenthesis, that 
Batatas, the scientific name adopted by Linnaeus, and as the 
name of a genus by Choisy, is the Spanish plural of Batata, 
the aboriginal name.) Even Peter Martyr and Oviedo do 
not agree in all particulars as to the distinction between 
"Ajes" and "Batatas" — a distinction which both recognize. 
In the 9th book of his 2d Decade, written about 1514, Peter 
Martyr (ed. 1574, p. 191), describing the fruits, etc., of the 
province of Uraba, Darien, names, for the first time, Batata?. 
" They dig from the earth," he says, " roots that grow spon- 
taneously (suapte natura nasccntcs') ; the natives call them 
' Batatas ' [accus. plural], which when I saw I thought to be 
rapes of Lombardy [? ' Insubres napos '] or great earth-tubers 
[ Cyclamen Europaium ? JZcqwm terras and Tuber terrce of the 
old botanists]. In whatever way they are cooked, roasted or 
boiled, they yield in delicate sweetness l to no confectionery 
or other eatable whatsoever." They are, he adds, " also 
planted and cultivated in gardens." In his 3d Decade (lib. 

1 The sweet potato was an inspiration to Peter Martyr, who rarely in- 
dulged himself in such a flight as " dulcorata mollities." 



ORIGIN OF CULTIVATED PLANTS. 319 

4, p. 240) lie mentions " Maize, Yucca, Ages, and Battatas " 
as plants that grew in Honduras when Columbus landed on 
that coast in 1502 ; and in the same Decade (lib. 5, p. 261) 
he names the same four plants as the ordinary food of the 
people of Caramaira (east of Darien) " as of the others," and 
again takes occasion to name the battatas, as surpassing all 
else " mira quad am dulci mollitie — especially if one falls on 
the better sort (nobiliores) of them." 

Oviedo gives a good description of the Batata, which, when 
he wrote (1525-35), was commonly cultivated by the Indians 
in Hispaniola and elsewhere, and highly prized (Historia, lib. 
vii. c. 4). It resembles the "Ajes," he says, in appearance, 
but tastes better and is far more delicate. The leaf is more 
notched (Jiarpadd) than that of the "Age," in nearly the same 
fashion. Some varieties are better than others, and he gives 
the names of the five kinds which are most highly esteemed. 
[Peter Martyr (dec. iii. lib. 9. p. 302) included the same five 
names among the nine varieties of " Ajes " that he mentioned 
as distinct ; but in this, as in other matters pertaining to natural 
history, Oviedo is the better authority.] " When the Batatas 
are well cured, they have often been carried to Spain, where 
the ships happened to make a quick passage, but more often 
they are lost on the voyage. Yet," adds Oviedo, " I have 
carried them from this city of Saint Domingo, in Hispaniola, 
to the city of Avila, in Old Castile." 

The " Gentleman of Elvas " who wrote the " True Rela- 
tion " of De Soto's expedition to Florida, in 1538, mentions 
Batatas, then growing in the Island of Terceira (belonging to 
Portugal). 

Cieca de Leon, who was in Peru in 1547, speaking of the 
fertility of the valle} T s near the Pacific coast, and the plants 
cultivated by the Indians, names among these, Sweet potatoes 
(Chron. del Peru, c. 66). In the Quichuan language they 
were called " apichu " ; in the dialect of Quito, " cumar." Mr. 
Markham, in a note to his translation (Hakluyt Soc, 1864, 
p. 234) mentions, on the authority of Dr. Seemann, " the 
curious and interesting fact that ' kumara ' is also the word 
for Sweet potato in Tahiti, the Fiji Islands, and New Zea- 



820 REVIEWS. 

land." Gavcilasso says these roots " which the Spaniards call 
'batatas' and the Indians of Peru 'apichu,' " are of four or 
five different colors, etc. " The least good are those that have 
been brought from Spain." 

Jean De Lery found them in Brazil in 1557, and described 
them under their Tupi name — " Hetich," as he wrote it — of 
which " the soil of Brazil is as prolific as that of Limousin or 
Savoy is of rapes." He describes the Indian method of plant- 
ing ; yet, " since these roots are the principal article of food 
of this country, and are met with by travelers in various 
places, I judge that they grow spontaneously " (Hist. Navig. 
in Brazil, p. 1G5). Montoya (Tesoro, 1G39) gives the Tupi- 
Guarani name, " Yeti" and mentions numerous varieties. 1 

Monardes, in the third part of his " Simpl. Medic, ex Novo 
Orbe," published in 1574 (translated by Clusius, ed. 1593, 
p. 439), states that Battatoe " are now so common in Spain, 
that ten or twelve caravel loads are sent annually from Velez- 
Malaga to Seville." 

De Candolle (who has elsewhere printed a short article upon 
the subject) calls attention to the fact, which ought to be fa- 
miliar, that sweet potatoes are roots, not tubers, and that 
Turpin long ago published good figures illustrating this ; also 
that while these roots are free from acrid or noxious qualities, 
all the Conrolvidacece with tubers, of which there are many, 
and not a few of lai-ge size, are inedible and acrid, — mostly, 
as we know, violently purgative. 

Manihot utilissima, Manioc, Cassava-plant. — De Candolle 
assigns good reasons for concluding (as did Robert Brown, 
without giving his reasons) that this important food-plant of 
the tropics is American, not African. But he leaves un- 
noticed the convincing fact that " Manioc " and " Manihot " 
are Brazilian names, slightly corrupted, of a plant cultivated 
in St. Domingo and Cuba before the landing of Columbus, 
and which became known to Spanish and Portuguese discov- 
erers before 1500, by its Haytian name, " yuca," or " hiucca." 

1 Hans Stade, who was a captive in eastern Brazil in 1549, briefly 
mentions these "roots called ' Jetlilci,'' of pleasant taste." (Captivity, 
Hakl. Soc. ed. p. 1G6.) 



ORIGIN OF CULTIVATED PLANTS. 321 

Peter Martyr (1493) describing the food of the islanders, 
names "Iucca, from which they make bread " (Dec. i. lib. 1, 
p. 7 ; ed. 1574) ; in the third book of his second decade 
(p. 148) he mentions Iucca, Ajes, and Maiz, as the three 
plants used by the natives for bread ; in the third decade 
(lib. 5, p. 262) he describes the mode of the propagation by 
cuttings, of cultivation, and of the preparation of " Cazabbi " 
from the root ; and he states that " there are many kinds of 
' iucca ' ' (p. 263). Oviedo (Historia, lib. 7, c. 2) describes 
"the bread of the Indians that is called ' cacabi,' " which 
is " made from a plant they call ' yuca,' " and he distinguishes 
two species of the plant. Acosta (Hist, of the Indies, transl. 
by E. G. ; Hakluyt Soc. ed., p. 232, 1588-90) gave a good 
account of the plant "yuca," and the kind of bread made 
from it, called " cacavi." 

Peter Martyr (Dec. iii. lib. 9, p. 301) relates the Haytian 
tradition of the origin of the cultivation of " yuca " in their 
island. " They say that a ' Boitius ' [i. e., magus, or diviner], 
a wise old man, after the lapse of many years, saw, on the 
banks of a river, a plant that was like a cane ; pulling it from 
the earth, he made this wild plant a cultivated one. He who 
first ate the ' Iucca' raw, quickly died. But because its taste 
was sweet, they determined that a way of using it should be 
diligently sought for. "When roasted or boiled, it was less 
hurtful. At last they came to the knowledge of the latent 
poison in its juice," etc. 

Gomara (Hist, gen., c. 71), Acosta (Hist. nat. y moral de 
las Indias, 1588-90 ; lib. 4, c. 17), Monardes (De Simplici- 
bus medic, transl. by L'Ecluse, 1593, p. 437), and other 
writers of the 16th century gave good descriptions of the 
plant " Yuca," and of the " cacavi " or " cazabi " prepared 
from the root. By the blunder of European editors, in the 
last half of the 16th century, the Haytian name was trans- 
ferred from the plant to which it belonged to one of another 
order, the Yucca of Linnaeus and of modern botany. The 
mistake was pointed out by Lobel. 

Jean de Lery (Hist. Navig. in Brasil. c. 9) describes the 
two species that were cultivated in Brazil in 1557 — under 



•22 REVIEWS. 



■ -- 



their Tupi names, " Aypi " \M. a#^i,Pohl] and " Maniot " 
[J/, utilissima]. Marcgrav (Hist, plant. Bras., p. 65) men- 
tions many varieties of both species, and gives " Mandioca " 
as the name of the root ; " Mandiiba " or " Maniiba " for the 
plant. Of the products of the root, Cassava retains its Hay- 
tian name (" cacavi ") nearly ; Tapioca is a corruption of the 
Brazilian (Tupi) "tipioca" or " tijnocui." 

Dioscorea sativa, alata, etc. Yam. — De Candolle informs 
us that these species, or their allies, are wholly unknown to 
botanists in a wild state : that, although cultivated in the 
East Indies, they have no Sanscrit names ; that they seem not 
to have been widely cultivated in Africa, but that the authors 
of the 17th and 18th centuries speak of them as widely dif- 
fused over the south Pacific islands, from Tahiti to New Cale- 
donia and the Moluccas. In the summary they are assigned 
to southern Asia (Malabar? Ceylon? Java?), and to the 
eastern Asiatic archipelago. Although a large part of the 
genus is indigenous to tropical America, it is thought that 
the cultivated species were probably introduced from the Old 
World. The following presentation of the evidence, as con- 
cerns America, may set the question in a different light : — 

The natives of Cuba and St. Domingo, when Columbus dis- 
covered those islands, cultivated two kinds of plants for their 
roots. These were called in the language of the islanders of 
St. Domingo, " Ages " or " Ajes," and " Yuca." Neither of 
these plants was known to the Spaniards. About " Yuca " 
there is no question ; it was the " Manihot," or " Manioc," of 
which we have already spoken. It is nearly as certain that 
the " Ages " was a species of Dioscorea, to which, in their ig- 
norance of the language of the islands, the Spaniards at first 
gave the name of "Name," " Niame," "Inhame," or other 
corruptions of a foreign (probably African) name ; and this 
name seems to have been occasionally misapplied both to the 
" Yuca " and the " Batata." 

L'Ecluse, who had traveled in the south of Spain and in 
Portugal, in 1563, says that the Colocasia (C. antiquorurri) 
" first brought from Africa, was common in many places in 
Portugal, near streams of water, that it was sought for by 



ORIGIN OF CULTIVATED PLANTS. 323 

negro slaves in Portugal, who ate it both raw and cooked," 
and that it was " called by the Portuguese, following the 
Moors, ' Inhauie,' — by the Andalusians, ' Alcolcaz,' ' ' etc. 
(Rarior Stirpium Hist., p. 299.) In a note to his transla- 
tion of " Garcia ab Horto " (1574, p. 217), he says that "the 
plant called ' Inhame ' by the Portuguese has very broad 
leaves, and grows near the water, or in water, — not spon- 
taneously, but when once planted it propagates itself from the 
roots," etc. 

Some of the companions of Columbus had seen the " In- 
hame " (or "Name ") in Africa, and were ready to transfer 
its name to the first cultivated roots they saw in America. A 
few days after the discovery of Cuba (Nov. 4, 1492), Colum- 
bus saw fertile fields " full of ' Mames ' [" these are ' Ajes ' or 
'batatas,'" notes Las Casas], which are like carrots (' Zana- 
horias '), and other plants, including Kidney beans and Beans 
(faxones y fabas) much unlike ours." (Navarrete, Colec, 
i. 200.) These "mames" are mentioned again Nov. 6 (id., 
203) — in both places, probably by an error of the copyist, 
for " niames " ; for, the next month, some natives of His- 
paniola brought " bread of ' niames,' which are roots that 
grow as large as rapes (' rtibanos '), which they plant and 
cultivate in all their fields, and on which they live ; and they 
make bread of them, boil them, and roast them ; and they 
have the taste of chestnuts, and no one eating them would be- 
lieve they were not chestnuts " (id., 238). A few days later, 
the Spaniards learned the names of these roots — or of others 
with which they were at first confounded. The Admiral sent 
a present to a friendly cacique. The officer who carried it 
reported, on his return, that " all this island ( St. Domingo) 
and Tortuga are cultivated like the country about Cordova. 
The lands are planted with ' Ajes ' — which are little shoots 
(ramillos) that are planted, and at the bottom of each grow 
roots like ' zanahorias ' which they use for bread," and these 
roots " are very savory, and taste like chestnuts." They have 
them here larger and better than he had seen in any place ; 
for, he said, he had " [seen] such also in Guinea " (id., p. 
242). Again, the natives "brought bread made of 'niames,' 



324 REVIEWS. 

which they call ' Ajes ' ' (id., 251) ; and, Dec. 2G, they gave 
the Admiral a "collation, of two or three kinds of 'Ajes,' 
and of their bread that they call ' cazavi,' " etc. (id., 263). 
After this the name of " niames" gives place to "ajes " (or 
" ages "). On the second voyage of Columbus, the natives, 
near Isabella (in St. Domingo), brought great quantities of 
" ages which are like rapes (nabos) very excellent eating," 
and " this age, the natives of Caribi (the Caribbean Islands) 
call nabi, and the Indians [of Hispaniola ?] hage " (id., 368, 
369). 

In two or three of the passages to which reference has been 
made — particularly those in which bread is mentioned — the 
Spaniards seem to have confounded the " ages " with the pro- 
duct of the " yuca " (Manihot), or to have included bothunder 
the general name of "niame" (or its equivalents, "Name, 
Igname," etc.) Amerigo Vespucci — or some one of the sev- 
eral translators through whom the relation of his first voyage 
comes to us — says, that in 1497, "the common food of the 
natives of Paria was the root of a certain tree (arborea radix 
quosdam), which they reduce to a good enough flour, and that 
some call this root ' Iucha,' others ' Cambi,' but others 
'Ignami'" (Navarrete, Colec, iii. 216). l 

This confusion of names, in the first decade of discovery in 
America, was natural and unavoidable. The foreign name, 
" niame, igname," was applied without much discrimination to 
roots cultivated by the natives of the islands and the mainland 
— primarily, to "ajes," occasionally to "yuca" (Manihot), 
and perhaps to " batatas." In the relations of the voyages of 
Columbus only two cultivated roots are named — " Ages " and 
" Yuca." The first book of Peter Martyr's first decade (dated 
1493, but probably revised before its publication in 1511) 
names only these two ; and in the third book of his second 
decade he mentions the use of the same two roots by the natives 

1 It is to this passage that Humboldt refers, in " Nouv. Esp.," 2d ed., 
ii. 468 (cited by M. De Candolle, p. 63), as evidence that the name 
" Igname " was heard on the continent of America by Vespucci in 1497 ; 
but, as will be seen, Vespucci (or his copyist) does not say that this name 
was used by the natives. 



ORIGIN OF CULTIVATED PLANTS. 325 

of Comagra, in Darien (p. 148) ; but in a subsequent chapter 
(dec. ii. c. 9., p. 191) be adds — as has been mentioned in a 
preceding- note — a third kind of roots, which the natives of the 
province of Darien call " Batatas," that grow in their coun- 
try spontaneously. From this date to the middle of the 16th 
century the distinction between these roots, though occasion- 
ally lost sight of, is generally observed. Oviedo (Historia, 
1. vii. cc. 2, 3, 4, pp. 268-73) describes the "ca^abi" and two 
species of the plant (" yuca ") that yields it ; " ajes " ; and 
" batatas." The " ajes," he says, were cultivated in Hispan- 
iola and in all the other islands, and on the continent ; they 
were of various colors — white, reddish, inclining to mulberry, 
and tawny, but all white within, for the most part ; the stem 
of the plant extends itself like that of " correhuela " (Con- 
volvulus or Bindweed), but stouter ; the leaves cover the 
ground, and are shaped much like " correhuela " and nearly 
like Ivy or panela, with some delicate veins (" unas venas 
delgadas "), and the little stems (" astilejos "), on which the 
leaves hang, are long and slender, etc. The leaf of the 
" Batata," he says (p. 274), is more toothed or notched (har- 
pada) than that of the " Aje," but of nearly the same fashion ; 
and the two plants are much alike, but the " Batatas " are 
sweeter and more delicate, etc. : some of the " Ajes " weigh 
four pounds each, or more. In some parts of Castilla del Oro 
(in Darien) there are "Ajes" that are small and yellow, etc. 
(p. 273). His description of the two plants permits no rea- 
sonable doubt that his " Ajes " were of the genus Dioscorea. 
Moreover, they were not identical with — though they resem- 
bled — the imported " name " or "yam": for Oviedo states 
(Historia, lib. vii. c. 19, p. 286), " that ' name ' (called 
' names ') is a foreign fruit, not natural to these Indies, which 
has been brought to Hispaniola and other places, and is suited 
to this evil race of negroes, and a profitable and good sub- 
sistence for them. . . . These ' names ' seem to be ' ajes,' 
but are not the same, and generally are larger than ' ajes.' ' 
They had already multiplied greatly in the islands and on the 
mainland. 

The distinction between " Ajes " and " Batatas," though 



32o REVIEWS. 

clearly apprehended, was sometimes lost sight of. Peter 
Martyr (dec. iii. lib. 9, p. 302) says that "the species of 
'Ages' are innumerable — the varieties being distinguished 
by their leaves and flowers ; " and he gives the American 
names of nine of the varieties ; but five of these nine are 
named by Oviedo (p. 274) as varieties of " Batatas." [See 
Batatas, ante.] 

The " Gentleman of Elvas," who wrote the narrative of 
DeSoto's expedition, mentions a fruit, at Santiago, Cuba, 
called " batata," the subsistence of a multitude of people, prin- 
cipally slaves, and which now (1538) grows in the island 
of Terceira, belonging to Portugal. ..." It looks like the 
' ynhame,' with nearly the taste of chestnuts " (Rela^am Ver- 
dadeira, ch. 5). 1 

Jean de Lery, who was in Brazil in 1557, though he gives 
a good description of the " Batata," does not mention the 
Yam ; but it is figured and described by Piso (Hist. Nat. 
Brazil., 1648, p. 93), as "Inhame" of St. Thomas, called 
" Cara " by the natives of Brazil, and " Quiquoaquecongo " 
by the Congo negroes. Ruiz de Montoya has the name 
'• Cara " in his Tupi dictionary, 1639, and mentions five vari- 
eties. As the Tupi name for the Virginia Potato (Sol ami m 
tuberosum), " Carati " Qi. e., white Yam), is formed from that 
of the " Inhame," it w r ould seem that the latter was of earlier 
introduction. So, in the Mpongwe — a language of the Congo 
group — the Potato is called " mongotanga," "white-man's 
yam." 

Portulaca oleracea, Purslain. — Botanists have taken it 
for granted that this weed of gardens and other cultivated 
grounds was transported to America from the Old "World. 
But Xuttall found it apparently indigenous on the upper Mis- 
souri forty years ago, and Dr. James in Long's Expedition, 
along the eastern base of the Rocky Mountains in what is now 
the State of Colorado. From thence to Texas it grows wild 

1 In one Indian language of the south, the Choctaw, the Sweet Potato 
is now called " ahe " ; while the Virginia Potato (<S. tuberosum) takes the 
adopted prefix of " Irish," " Ilish ahe," or is sometimes called " ahe lumbo," 
" round ahe." 



ORIGIN OF CULTIVATED PLANTS. 327 

along with two other nearly related species. Moreover, the 
following evidence tends to show that its introduction, if in- 
troduced by human agency, took place before the landing of 
Columbus. 

On their first sight of the New "World, the Spaniards were 
much impressed by the strangeness of all forms of animal and 
vegetable life : " all the trees are as unlike ours, as day is to 
night" — wrote Columbus, Oct. 17th, 1492, six days after 
landing at San Salvador : " and so are the fruits, and so the 
plants, and the stones, and all things" (Xavarrete, i. 183). 
On the 28th, on the north shore of Cuba, he saw — apparently 
for the first time — a familiar plant : " hallo verdolagas 
muchas y bledos," — he found much "purslane" and "ble- 
tum " (id., 192). It seems hardly possible that the Admiral 
and his companions could mistake a strange plant for a salad 
herb so well known as " verdolagas " to Spanish eyes and 
palates. Again, Oviedo, writing about 1526, in a list of 
"plants in the island of Hispaniola which are like those of 
Spain, and which were before the Christians came to these 
parts, and arc natives of this land, and were not brought from 
Spain," mentions " verdolagas or pertulaca," and " bledos or 
bletum" (Blitum). 

In his description of " perebenecuc," written in 1525. he 
says that plant grew in great abundance in Saint Domingo 
and in many places on the continent, in the woods and fields ; 
even "purslane ('verdolagas') is not more abundant here " 
(id., lib. xi. c. 5, p. 378). 

Jean de Lery, in Brazil in 1557, was as much impressed 
by the novelty of the flora as Columbus had been in the 
"West Indies. " I declare," he wrote (Hist. Navig. Brazil., 
168), " as far as it was permitted me to discover in wander- 
ings through the woods and fields, that there are no trees or 
plants, or any fruits, that are not unlike ours, these three 
excepted, portulaca, ocymum, and filex " (in the original 
French edition, 1578, p. 217, "pourpier, basilic, et fougiere"). 

Capt. John Smith, in Virginia in 1606, found "many herbes 
in the spring, commonly dispersed throughout the woods, good 
for broths and sallets, as Violets, Purslain, Sorrell, etc. ; be- 



328 REVIEWS. 

sides many we used whose names we know not " (Smith's 
Gen. History, 1632, p. 26; and repeated by Straehey, 
Travaile into Virginia, p. 120). Smith's purslain was 
probably Sedum tematum. 

Sagard-Theodat, in the relation of his " Grand Voyage dn 
Pays des Hnrons," in 1624 (p. 331), says that the Hurons 
make little use of herbs, " although the pourpier or pource- 
laine is very common there, and grows spontaneously in their 
fields of corn and pumpkins." 

W. Wood, who was in New England from 1629 to 1633, 
names " Purselane " among plants growing " in the woods, 
without either the art or the help of man " (N. E. Prospect, 
pt. 1, c. 5). We doubt its growing literally in the woods, as 
unlike its natural habit, and place more confidence in the 
statement of Champlain, who, in his earlier voyages, 1604-11, 
found plenty of excellent " pourpier," for his salads, on the 
coast of New England, growing among the Indian Corn ; " the 
savages making no more account of it than if it were a nox- 
ious weed " (Voyages, ed. 1632, p. 80). 

Humulus Lupulus, Hops. — Although the matter has noth- 
ing to do with the introduction of Hops into cultivation, it is 
noticeable that De Candolle assigns the home of the plant 
only to Europe and western Asia. It is undoubtedly indige- 
nous to North America also, and is mentioned as such in the 
American works. In Gray's " Manual," besides the printing 
of the name in the type appropriate to indigenous species, the 
plant is expressly stated to be " clearly indigenous." But, 
through some oversight, in the " Prodromus " (xvi. 29), it is 
stated, in connection with this very reference, that the plant 
was introduced. 

Oca. — Considering that " Mate " and " Coca " find place 
in this volume, although perhaps rather employed than culti- 
vated (at least the former), the absence of Oca QOxalis tuber- 
osa and O. crenata) is noticeable. This esculent root deserves 
mention, if only for the antiquity of its culture in Peru. The 
name, which is Quichuan, appears to have belonged, specially, 
to Oxalis tuberose/. Another root " like the oca in shape, but 
not in taste," called in Quichua " anus," was less esteemed. 



ORIGIN OF CULTIVATED PLANTS. 329 

Both were cultivated in Peru in the time of the Incas, and in 
the districts where no Maize grew, the crop of these tubers 
was of much importance (Garcillaso, Comment., b. v. c. 1 ; 
b. viii. c. 10). J. de Acosta, 1588-90, says "there are an 
infinite number " of roots used for food in the Indies, " but 
the Papas (Potatoes) and Ocas be the chief for nourishment 
and substance " (Nat. and Moral Hist, of the Indies, lib. 
iv. c. 18). 

Our notes upon plants cultivated for their herbage, tubers, 
roots, etc., have run to such a length that the remainder, con- 
cerning some plants cultivated for their fruits and seeds, must 
be left for another article. 



PART II. 

The fourth chapter relates to plants cultivated for their 
fruits ; the fifth, to those cultivated for their seeds. Our 
present annotations concern a few species or forms of Cucur- 
bitacece, the history of which has been involved in some ob- 
scurity and confusion. 

A word, in passing, upon the Peach, upon the history of 
which this volume throws some new light. De Candolle had 
formerly suggested China as its home, and he has brought 
together additional evidence in favor of that view. He 
shows how this conclusion goes against an old idea that the 
Peach is a derivative of the Almond, which is indigenous to 
western Asia, and was unknown to the Chinese anterior to the 
Christian era, while they had peaches of various sorts long 
before. Upon Pyrus there is a note relating to botanical 
orthography, p. 183, which we append, as it has an applica- 
tion to a few other words. 1 

1 " L'Orthographe Pyrus, adopte par Linnd, se trouve dans Pline, 
'Historia,' ed. 1631, p. 301. Quelques botanistes ont voulu raffiner en 
e*crivant Pirns, et il en result, pour une recherche dans un livre moderne, 
il faut consulter l'index dans deux endroits, ou risquer de croire que les 
Poiriers ne sont pas dans l'ouvrage. En tous cas le nom des anciens est 
un nom vulgaire, mais le nom vraiment botanique est celui de Linnd, 
fondateur de la nomenclature adoptde, et Linne - a ecrit Pyrus." 

Pears and apples were prehistoric in Europe, both wild and cultivated. 



n 



30 REVIEWS. 



Lagenaria vulgaris. Bottle Gourd. — Although doubtless 
an Old-World plant (De Candolle attributes it to India, Mo- 
lucca. Abyssinia), yet it is not quite certain that it had not 
reached the New World before Columbus. At least the 
following notes may be put upon record. 

M. De Candolle mentions the case of the name s:ourd for 
pumpkin (" potiron ") by English writers, as " an example 
of the confusion of popular names and the greater precision 
of scientific names." Such confusion becomes more perplex- 
ing when we have to deal with popular names of the 15th and 
16th centuries. Parkinson — a good observer and a respect- 
able botanist — complained, in 1640, of " our modern writers 
who confound Pepo, Melopepo, and Cucurbita so promiscu- 
ously that it is not possible to find out the distinct certainty 
of them all ; for some make that Pepo that others call Melo- 
pepo, and others, Cucurbita. " (Theater of Plants, p. 770.) 
Scientific names of the 16th century are as obsolete as pop- 
ular names of the same period. They do not help us to dis- 
tinguish Lagenaria from Cucurbita, or Pepo from Melopepo ; 
or Citrouille from Citrullus. Early voyagers to America 
wrote cucurbita, calabaca, courge, or zucca, as a name for 
any gourd or pumpkin, and occasionally for a calabash which 
was not even a cucurbit. The relation of the first voyage 
of Columbus repeatedly mentions the " calabazas " used by 
the natives of St. Domingo and other islands for carrying 
water (Navarrete, Collec, i. 180, 188, and Dec. 3, 1492). 
Columbus saw, near the east end of Cuba, fields planted with 
" calabazas" and other productions of the country (id., p. 225). 
Yet we know from Peter Martyr that some of the gourds 
(" cucurbitse ") used in the islands grew on "cucurbiteas 
arbores " as tall as elms (Dec. i. lib. 3, and iii. lib. 4 ; pp. 
38, 246). This tree, Crescentia Cujete, is described by 
Oviedo (Historia, lib. viii. c. 4) under its Haytian name, 
" Higuero " ; in Nicaragua it was called " Guacal " ; and in 
Brazil, " Cuiete " l (Marcgrav, Hist. Nat. Brasil., 123). J. de 

1 Not " Cujete " — unless y has the German sound. The Tupi name is 
formed from cu'i (cou-in, Lery) " the shell " or hard rind of a nut or fruit 
(and the bowl or calabash made from it) and etc " good, precious." 



ORIGIN OF CULTIVATED PLANTS. 331 

Lery (Hist. Navig. in Brasil., 154) describes the tree under 
the Brazilian name of " Choyne " ; but elsewhere (p. 246) 
he says "the natives have Cucurbitce (courges) and other 
kinds of fruits," from which "they make their bowls, called 
4 coui,' and other vessels." 

It is certain that "calabacas," which were not arboreal, 
but genuine cucurbits, were abundant — and were believed to 
grow spontaneously — in the islands and on the main land, 
before 1526. Oviedo (Historia, lib. vii. c. 8) observes that 
" calabacas, in the Indias, were as common as in Spain, and 
of the same kinds (de las mismas), long and round, or banded 
(eenidas), and of all the shapes they usually have [in Spain]." 
They were much used " in all parts of these Indias, both the 
Islands and the Main," and " are one of the common things 
that the Indians cultivate in their gardens." They were not 
cultivated for food — " for they do not eat them " — but for 
carrying water ; " and they have other calabacas that are 
in all respects like the aforesaid, except that they are bitter 
to the' taste; and there are many of these that grow of them- 
selves without cultivation." : The same author (lib. xi., e. 1), 
in a list of plants introduced from Spain, names Melons and 
Cucumbers (pepinos), but not Gourds. 

The relation of the voyage of Amerigo Vespucci, 1489, in 
a description of the Indians of Trinidad and the coast of 
Paria, says that " each carried, hanging at his neck, two 
small dried gourds (cucurbitas), one containing the plant 
that they were accustomed to chew, the other, a certain 
whitish flour," etc., and that each woman carried a " cucur- 
bita" of water (Navarrete, iii. 252, 254). 

The " Cucurbita lagenae forma," which Marcgrav found 
in Brazil, 1637-8 (Hist. Nat. Brasilia?, 44), though "very 
probably Lagenaria vulgaris" yet, as M. De Candolle ob- 
serves, " does not prove that the species was in that country 
before the voyage of Amerigo Vespucci in 1504 ; " but we 

1 M. De Candolle, p. 198, citing this passage from Ramusio's Italian 
translation of Oviedo's " Historia," has " zueche " for " calahacas " of the 
Spanish original, and takes no notice of what is said of their spontaneous 
growth. 



332 REVIEWS. 

know from Lery, above cited, that the natives of Brazil used 
Cucurbitce, for bowls and drinking-vessels, at least as early 
as 1557. Moreover, the richness of the Tupi vocabulary in 
gourd-names suggests — if it does not absolutely prove — that 
several varieties of Lagenaria were known to the Brazilians 
long before the visit of Piso and Marcgrav. The " Tesoro de 
la Lengua Guarani " (o Tupi} of Father Ruiz de Montoya 
was first printed in 1G39. It gives for gourd ("calabaco"), 
the Tupi general name, " la " [which is a compound of 
t " water," and yd or a " fruit "] , and for the varieties — among 
others — it names tact " round gourd " ; idyurumi " narrow- 
mouthed gourd " ; tail " long-necked gourd " ; idobd " wide- 
mouthed gourd " ; mquatia " painted gourd " ; idcuipe, " spoon 
gourd " (used for making spoons) ; idape " small gourd, used 
for drinking " ; idqaa " great gourd " ; idcui " gourd like a 
great dish " or bowl, etc. : not including the derivatives of cm, 
or the edible " calabacas " — to be mentioned hereafter. 

" Acosta, too," says M. De Candolle, " speaks of Cale- 
basses which the Peruvians used for cups or vases, but the 
Spanish edition of his book is of 1591, more than a hundred 
years after the conquest." (?) Acosta says more than this. 
After mention of the " Calebasses or Indian Pompions . . . 
especially those which ai-e proper to the country" [Peru], he 
adds : " There are a thousand kinds of Calebasses ; some are 
so deformed in their bigness that of the rind cut in the midst 
and cleansed, they make, as it were, baskets to put in all their 
meat for their dinner. Of the lesser, they make vessels to 
eat and drink in," etc. (Hist. nat. y moral de las Indias ; 
translation, revised by Markham, lib. iv., c. 19, p. 238.) 

Cucurbita maxima, C. Pepo, C. moschata. Pumpkin, 
Squash, etc. — In the " G6ographie Botanique " not one of 
the cultivated Cucurbitce is attributed to America, and a ref- 
erence to Nuttall's record that the warted squash was grown 
by the Indians on the upper Missouri is the only mention of 
any aboriginal cultivation of squashes in North America. In 
the present volume there is merely a reference, in this re- 
spect, to Dr. Harris's article in this Journal (xxiv., 1857), 
and to Mr. Trumbull's note in the " Bulletin of the Torrey 



ORIGIN OF CULTIVATED PLANTS. 333 

Club " (1876), with the comment that : " Cela nous apprend 
seulement que les indigenes, un siecle apres la decouverte 
de la Virginie, 20 a 40 ans apres la colonisation par W. 
Raleigh, faisaient usage de certains fruit de Cucurbitacees." 
Nevertheless Cucurbita Pepo, upon botanical indications 
solely, is attributed to temperate North America in the gen- 
eral table, to a Mexican or Texan origin in the body of the 
work. This rests upon the collection by Lindheimer, in 
Texas, of a form of this species " apparently indigenous." 
That was between thirty and forty years ago ; no wild speci- 
men has since been received from all that region (nor from 
any other) ; and it is w T ellnigh certain that the species was 
commonly cultivated in all that country by the aborigines. 
If ever found truly indigenous, it will probably be farther 
south than Texas. C. maxima is now set down as from 
Guinea, on the strength of a single finding of it " apparently 
indigenous " on the banks of the Niger. C. moschata (to 
which Vilmorin refers the Canada Crook-neck Squash) is in 
the list of species of completely unknown or nncertain origin. 

In this state of the case, it is certainly worth while to pre- 
sent the evidence — gathered with much care and pains — 
which assures us that one or two, and perhaps all three, of 
these species, and many varieties, were largely cultivated 
throughout America, from the tropics to Canada, before the 
voyages of Columbus. 

Allusion has already been made (under Lagenaria) to the 
difficulty of distinguishing the genera of Cucurbitacece under 
the names by which they are mentioned by voyagers and 
explorers of the first century after the discovery of America ; 
and the question of species is particularly difficult. Yet we 
find abundant evidence — especially as respects North America 
— (1) that, in various parts of the country remote from each 
other, the cultivation of one or more species of Cucurbits by 
the Indians was established before those places are known to 
have been visited by Europeans ; (2) that these species or 
varieties were novel to Europeans, and were regarded by bot- 
anists of the sixteenth and seventeenth centuries, as well as by 
the voyagers and first colonists, as natives or denizens of the 



334 REVIEWS. 

region in which they were found ; and (3) that they became 
known only under American names ; one of these names 
(Squash) becoming, in popular use, generic, and two others 
(M acock and Cushaw) surviving, as names of varieties, into 
tbe present century. 

To present this evidence as nearly as possible in the order 
of time, we refer, first, to the relation of the first voyage of 
Columbus. December 3, 1492, entering a small river [the 
Rio Bonia], near the eastern end of the island of Cuba, he 
found near it a populous Indian village, and saw large culti- 
vated fields " planted with many things of the country, and 
calabazas, a glorious sight (que era gloria vella) ! " See 
Navarrete, Colec, i. 225. It is not certain that these " cal- 
abazas" were not bottle-gourds (Lagenaria), but it is, to say 
the least, highly inrprobable that the enthusiasm of Columbus 
would have been so kindled by the promise of a harvest of 
little value to Europeans. 

Oviedo (Historia, 1. xi. c. 1) names among plants and seeds 
brought from Spain to Hispaniola " melones " and " pepinos " 
— of which imported varieties were already abundant in the 
island before 1535 ; the seed of " cogombros " brought from 
Castile had not succeeded so well. 

In July, 1528, Cabeca de Vaca found near Tampa Bay, in 
Florida, "maize, beans, and pumpkins in great plenty, and 
beginning to be fit for gathering." In 1535-6, when passing 
through Texas, the Indians supplied him with prickly pears 
and, occasionally, maize ; but after crossing " a great river 
coming from the north ' ' — probably the Rio Grande — he 
and his companions came to a region having " fine dwellings 
of civilization, whose inhabitants lived on beans and pump- 
kins" — and, when the season was not too dry for raising it, 
maize (Relacion, 1542 ; translated by B. Smith, 1871). 

In the summer and autumn of 1539, De Soto found the 
Appalachian country, in western Florida, well supplied with 
" maize, beans (fesoles) and pumpkins (calabaeas) ; " the 
pumpkins of Uzachil were " better and more savory than 
those of Spain ; " there were " fields of maize, beans, and 
pumpkins," not far from Tampa Bay, where he first landed 



ORIGIN OF CULTIVATED PLANTS. 335 

from Cuba ; at Pacaha, on the Mississippi, the northernmost 
point he reached (1541), he found again " many pumpkins 
and much maize and beans " ; and, still westward, at Coligoa, 
" beans and pumpkins were in great plenty ; both were larger 
and better than those of Spain ; the pumpkins when roasted 
had nearly the taste of chestnuts " (Oviedo, lib. xvii. cc. 24, 
28 ; True Relation, etc., by a Fidalgo of Elvas ; translated 
by Buckingham Smith, pp. 45, 47, 122, 285). Oviedo writes 
" calabacas," but the author of the Portuguese " Relacani 
Verdadeira " (1557) has, in one or more of the places cited, 
" aboboras." 

In 1535, Jacques Cartier, the first explorer of the St. Law- 
rence, found among the Indians of Canada " grand quantite" 
de gros melons, concombres and courges " (Bref Recit de la 
Navigation, etc., 1545 ; reimpr. Tross, 1863, ff. 24, 31). 

Sagard, whose " Grand Voyage du Pays des Hurons " was 
made in 1642, makes repeated mention of the native squashes 
(" citrouilles du pays") which the Hurons raised in abun- 
dance, and which he found very good, boiled or baked (pp. 85, 
105, 140, 331). In his " Histoire du Canada " (283) he 
describes the method by which the Indians hastened the ger- 
mination of the seeds of these " citrouilles du pays," and 
"raise them with great ease." 

Lahontan (Xouv. Voyages, 1703, ii. 61) describes the 
" Citrouilles " of (southern) Canada — " sweet, and of a dif- 
ferent kind from those of Europe, where," as several persons 
assured him, these would not grow. " They are of the size 
of our melons; the flesh yellow as saffron. They usually 
bake them in the oven, but they are better roasted under the 
embers, Indian fashion," etc. Lahontan had as little doubt as 
Sagard had, that these " citrouilles " (cultivated by the Indians 
of Canada from the time of Cartier, at least) were genuinely 
" du pays." 

As to the Cucurbitacece of Virginia, M. De Candolle admits, 
" only, that the natives, a century after the discovery of Vir- 
ginia, twenty to forty years after the colonization by ~Y\ . 
Raleigh, made use of certain fruits of Cucurbitacece " 
(p. 201). Let us reexamine the evidence. Captains Amidas 



336 REVIEWS. 

and Barlow, in the first vessels sent by Sir Walter Raleigh to 
the New World, landed on an island in Ocracoke Inlet (now 
within North Carolina) in 1584. While the vessels remained 
there, and while they were at Roanoke Island near by, the 
Indians entertained them kindly, and " sent them, commonly 
every day, a brace of bucks, conies, etc., sometimes melons, 
walnuts, cucumbers, pease, and divers roots " (J. Smith's 
Gen. Hist. p. 3). 

What these " melons," or some of them, were, we learn 
from later explorers and the first colonists of Virginia 
(proper). 

Captain John Smith says that the Indians of Virginia 
(1606-8) " plant amongst their corn Pumpions, and a fruit 
like unto a musk-melon, but less and worse, which they call 
Macocks," etc. (Gen. Hist. p. 29). Strachey, who was in 
Virginia in 1610, describes these " macock gourds " in nearly 
the same words (Trav. into Virginia, p. 72) ; elsewhere, he 
says the " macokos is of the form of our pumpions — I must 
confess, nothing so good, — 't is of a more waterish taste," and 
he mentions also the " pumpions " planted by the Indians, 
and " a kind of million " which they " seeth and put into 
their walnut-milk, and so make a kind of toothsome meat " 
(p. 119). " The Indian Pumpion, the Water-melon, Musk- 
melon," etc., are named among fruits introduced into Ber- 
muda, by the English, before 1623 (Smith's Gen. Hist, 
p. 171). 1 

1 L'Ecluse (Clusius) heard of these Macocks in 1591 or earlier. In 
his "Exotica" (1G05 ; lib. iii. c. 2) he describes a fruit — " Macocqwer 
Virginiansium, forte " — which had been sent him from London by James 
Garet, brought from " the province of Wingandeeaow, which the English 
call Virginia." He conjectured that this might be " the fruit which the 
natives of that region call Macocqwer" — but his figure and description 
do not favor this identification. The fruit, he says, is nearly orbicular ; 
four inches in diameter ; with a hard rind, yellowish on the outside ; many 
seeds, flat and heart-shaped (" cordis, ut vulgo pingitur, formam referen- 
da "). L'Ecluse thought it might be one of the gourds which the natives 
used for rattles, as the Brazilians used their Tamaraca, etc. His speci- 
men was old and dried, the pulp blackened, the rind covered with a dark 
membrane, "perquam sparsa? quaedam fibra3 a pediculo ad summum." 
This must have been a fruit of Crescentia cucurbitina, a calabash, which is 
a native not only of the West Indies, but also of southern Florida. 



ORIGIN OF CULTIVATED PLANTS. 337 

Among Johnson's additions to Gerarde's " Herball," 1636, 
there is a description of " Macock Virginiani, sive Pepo Vir- 
ginianus ; the Virginian Macock or Pompion " (pp. 919, 
921). The description is dated 1621, and signed by John 
Goodyer. The plant has " great broad shrivelled yellow 
flowers, like those of the common Pompion.'' 1 The fruit, "'some- 
what round, not extending in length, but fiat like a bowl, but 
not so big as an ordinary bowl, being seldom four inches broad 
and three inches long ; of a blackish green color when it is 
ripe. The substance or eatable part, of a yellowish white 
color. . . . Seeds like the common Pompion, but smaller." 
The "small round Indian Pompion," and "the cornered In- 
dian Pompion " — the latter resembling our common " scol- 
loped Squash " (" Pepones lati, Broad Melons or Pepons " of 
Lyte's Dodoens, p. 588) — are described and figured in " John- 
son's Gerarde," p. 920. 

Beverley's " History of Virginia," 1705, p. 124, mentions 
the Macocks, " a sort of Melopepones, or lesser sort of Pom- 
pion or Cashaw," which he identifies with the " Squash or 
Squonter Squash " of New England. " The Indian name," he 
says, " is still retained by them." Professor Scheie de Vere 
(of Virginia) states that it still " survives in its anglicized 
form of Maycock" (Americanisms, 1871, p. 60). 

The " Cushaw" ("Ecushaw," Hariot) is described by Bev- 
erley (Hist, of Virg., p. 124) as " a kind of Pumpion, of a blu- 
ish green color, streaked with w r hite when they are fit for use. 
They are larger than the Pompions, and have a long narrow 
neck. . . . The Cushaws and Pompions they lay by, which 
will keep several months good, after they are gathered " 
(p. 152). Bartlett, " Diet, of Americanisms," notes the name 
Cushaw, " sometimes spelled Kershaw," as " Western " for a 
pumpkin. Beverley's description makes it nearly certain that 
the variety so named was the (New England) winter " crook- 
neck " squash — which, five and twenty years ago, might have 
been seen hanging, by its necklace of flannel " list," in every 
New England farmer's kitchen, from early harvest time till 
wanted for Thanksgiving or Christmas pumpkin-pies. 

The Rev. Francis Higginson, who came to New England in 



338 REVIEWS. 

1 629, wrote from Salem, a few weeks after his arrival : " Here 
are stores of pompions, cowcumbers, and other things of that 
nature which I know not " (N. E. Plantation, 1G30) ; and, 
again: "We abound with . . . sundry sorts of fruits, as 
musk-melons, water-melons, Indian pompions, Indian pease, 
I nans, and many other odd fruits that I cannot name" 
(Young's Chron. of Mass., 265). William Wood, who was 
in New England from 1629 to 1633, says of the Indians of 
Massachusetts: "In summer, when their corne is spent, 
' Isquoutersquashes ' is their best bread, a fruit like a young 
Pumpion " (N. E. Prospect, p. 76). Roger Williams, 1643, 
names these " Askutasquash, their vine-apples, which the 
English from them call squashes, about the bignesse of apples, 
of several colors, a sweet, light, wholesome refreshing" 
(Key to the Language of America, 103). Again, Josselyn 
(1638-71, N. E. Rarities, 57) mentions these " squashes . . . 
more truly Squontersquashes, a kind of melon or rather gourd, 
for they oftentimes degenerate into gourds ; some of these 
are green, some yellow, some longish like a gourd, others round 
like an apple, all of them pleasant food boiled and buttered, 
etc. But the best yellow squash, called an Apple squash, 
because like an apple, and about the bigness of a Pome-water, 
is the best kind : they are much eaten by the Indians and the 
English." But he distinguishes these from the " Pompions 
[of which] there be several kinds, some proper to the coun- 
try ; they are dryer than our English Pompions, and better 
tasted ; you may eat them green " (p. 91). The last words 
(here italicized) give a nearly literal translation of the 
Alironkin-Indian name of Cucurbits, — in the dialect of New 
England, " asq," plural " asquash," " green things," or (to be 
eaten) " immature." Eliot, in his version of the Bible (1663) 
names three kinds of asquash : askoot-asquash [ = Askut- 
asquash, R. Williams, Isquoutersquash-es of Wood, Squonter- 
squash-es of Josselyn, ut supra], for "cucumbers"; quonoo- 
asquash " gourds " [literally, " long asquash "] ; and monas- 
koot-asquash " melons." 

Squashes were first known to the Dutch by their Al- 
gonkin name. Van der Donck, after speaking of the pump- 



ORIGIN OF CULTIVATED PLANTS. 339 

kins of New Netherland (1642-53), adds : " The natives have 
another species of this vegetable peculiar to themselves, called 
by our people ' quaasiens,' a name derived from the aborigines, 
as the plant was not known to us before our intercourse with 
them. It is a delightful fruit, as well to the eye on account 
of its fine variety of colors, as to the mouth for its agreeable 
taste. ... It is gathered early in summer, and when it is 
planted in the middle of April, the fruit is fit for eating by 
the first of June. They do not wait for it to ripen before 
making use of the fruit, but only until it has attained a certain 
size. They gather the squashes and immediately place them 
on the fire without any further trouble. . . . The natives 
make great account of this vegetable." (Descript. of N. Neth- 
erlands, 1656 ; transl. in N. Y. Hist. Soc. Coll., 2 ser., i. 186.) 

Thus far we have cited, with one or two exceptions, Amer- 
ican authorities. M. De Candolle, after mentioning " the 
three forms of Pepones figured by Dodoens, edition of 1557, 
to which a fourth, P. rotunaus major, was added in the edi- 
tion of 1616," and a figure of P. oblongus, iu Lobel. " Icones," 
641, observes, that " the names given to these plants indicate 
a foreign origin ; but the authors can affirm nothing in this 
regard ; the less so, because the name Indian signifies, either, 
of southern Asia or of America" (p. 204). A collation of 
the descriptions of Pepones or Cucurbitse, given by European 
botanists of the sixteenth century, does away with this am- 
biguity. 

Tragus (Hieron. Bock), " De Stirpium Xomenclaturis," 
etc., 1552, p. 830, described and figured Melo, Pepo, Cucumis, 
and Citreolus ; and (p. 832) named, also, Cucumis sylves- 
tris. In the next chapter (p. 834) he wrote "De Cucumere 
seu, ut vulgo loquuntur, Zucco marino " — with a figure. 
" Many kinds of strange plants," he says, " have been brought 
from remote parts into Germany, in the last few years." 
Among others, these " poma oestiva," of which some are large, 
some small, some round, some oblong, some sweet, others 
bitter, of various colors. " Some call these ' Cucumeri,' and 
assert that they are ' Turkish Cucumeres,' with which opinion 
I cannot agree. ... I call them Mala cestiva & Jtidica" of 



340 REVIEWS. 

which lie distinguishes four kinds, 31. Indica erocea, lutea, 
citrina, and nigra. " Commonly," he says, u they are called 
Zucco marina, because they first came to us from parts 
beyond the sea, some from Syria, some from India, which the 
names given them attest ; for they are commonly called Zucco 
de Syria and Zucco de Peru." 

The figure of Cue timer marinus, Ital. Cocomere marino, 
etc., in the " Effigies Plautarum " of Fuchs, 1549, is a reduced 
copy of Bock's, and substantially agrees with that of Pepo 
rotundus in Lyte's "Dodoens," p. 587, which was "called 
also Cucumis marinus ; of some, Zucco marino ; in French, 
Concombre marin, Pompons Turquins" etc. 

Matthioli of Padua (Comm. in Dioscor., ed. 1559, p. 292) 
is more explicit. " There are," he says, " various kinds of 
cucurbits foreign to Italy, which can be kept fresh far into 
the winter. They say that these came into Italy from the 
West Indies, whence they are called by many Indian. Their 
taste is sweetish, not so insipid as ours," etc. ; and his figure 
of Cucurbita Indica agrees with that of Bock's Zucco mari- 
nus (or " Zucco de Peru ") and with Lyte's Pepo rotundus. 

It is certain, then, that the botanists of the 16th century to 
whom M. De Candolle refers, used Indian — when applied to 
varieties of Cucurbita — in the sense of American. In the 
17th century, the evidence is not less direct. Parkinson 
(Theatrum Botanicum, 1640, pp. 769, 770) figures and de- 
scribes (1) Cucurbita lagenaria major, the greater Bottle 
Gourd ; (2) C. longer, the long Gourd : (3, 4) C. clypeifor- 
mis & verrucosa, and Anguria jSHgyptiaca, the Simmel 
[Scallop Squash], and the rugged Gourd [warted Squash, 
orbiculate depressed], and the Egyptian Citruell or Watery 
Million : (5) Cucurbita Indira* oralis, pyriformis, & fere 
rotundus* Indian Gourds, oval, pear-fashioned, and almost 
round. Of these Indian Gourds he says : " There is very 
great variety of these Gourds (or Millions, as some call them, 
or Pompions, as I may call them) that came out of America 
or the West Indies, from sundry places, both farther south 
among the Spanish colonies, and nearer hand, in our own of 
Virginia, New England, etc." He notes the great variety 



ORIGIN OF CULTIVATED PLANTS. 341 

of size, shape, and color, " some as great as our pompions, 
some as small as an apple, some discolored on the outside, 
green with whitish or yellowish stripes, . . . some also reddish, 
spotted, or striped, and some of a deep yellow." 

Piso and Marcgrav (Hist. Nat. BrasiL, 1648, p. 44) de- 
scribe and figure a plant called " Iurumu " [=Yunimu] by 
the Brazilians, and by the Portuguese, "Bobora." M. De 
Candolle, p. 201, is inclined to agree with modern botanists in 
referring this to C. maxima; but, as he remarks, it appears 
to have been a cultivated plant. If introduced from abroad, 
the name given it by the Tupis was probably formed, by pre- 
fix or affix, from that of some native (or naturalized) species 
to which it had some resemblance. In Montoya's "Tesoro," 
1639, we find"Yurua" " calabacillos silvestres," small wild 
calabazas; but the name "Yurumu" did not yet appear. 
Almost a century before the visit of Piso and Marcgrav, Jean 
de Lery saw in Brazil (1557) " certains citrouilles rondes, 
fort douce a manger," called by the natives " Maurongaus " 
(Voyage, ed. 1578, p. 217). The Tupi name "moranga" 
(the first two vowels nasal) denotes a " handsome fruit." 

Lycopersicum escidentiim, Tomato. — We have only to note 
an oversight in respect to the Mexican cultivation of the Mala 
Peruviana, as it was named by some botanists of the 16th 
century. De Candolle refers to Humboldt's statement that 
the cultivation of this esculent was ancient in Mexico, but 
adds that there is no mention of it in the earliest work on the 
plants of that country, namely : Hernandez, " Historia." But 
Hernandez (ed. 1651, p. 295 ff.) actually has a chapter " De 
Tomatl, sen planta acinosa vel Solano," and describes several 
sorts under their Mexican names. 1 

1 I find only one writer in the 16th century who gave the Tomato a 
name indicating 1 a Peruvian origin — namely, Anguillara, whose treatise 
" De Simplicibus " was first printed in Italian, at Venice, 1561. On his 
authority the name Poma Peruviana is introduced in the synonymy by C. 
Bauhin, in his annotations on Matthioli, 1598 (p. 761), and the "Hortus 
Eystettensis " (attributed to Besler), published in 1613, is referred to by 
the same writer (Pinax, 1771, p. 167) an authority for the names, inter alia, 
" Poma amoris fructu rubro & Mala Peruviana." The "Hortus Eystet- 
tensis " is the only authority cited for Mala Peruviana in the work to 



342 REVIEWS. 

Persea gratissima, Alligator Pear of the English, 1' Avoeat 
of the French ; a singular corruption of a native name, as De 
Oandolle remarks, which had no more to do with an alligator 
than with a lawyer. Our author does not carry back the 
native Mexican name quite to its original, which was " Ahua- 
cahuitl " corrupted by the Spaniards into " Aquacate," 
" Avogade," etc. Champlain, who saw it in Mexico in 1599 
or 1600, calls the fruit " Accoiates " and " Acoyates " ; " Voy. 
to the W. Indies"' [Hakluyt Soc, 1859], p. 28. 

Oviedo described " the wild pear-tree of the main land," in 
1526. It grew " in the province of Castilo del Ora (Panama), 
in the sierras of Capira and the country of the cacique of 
Juanaga," etc. In the revision of his first work, in 1535, he 
adds, that he had, some years before, seen these trees culti- 
vated by the Indians in Nicaragua (Historia, lib. ix. c. 22). 
It was still a tree of " Terra firma " — not yet introduced 
into the Islands. Clusius saw it in a garden in Valencia — 
" said to be brought from America " — thirty-five years ear- 
lier than the date (1601) mentioned by De Candolle. He 
described the Persea in the first edition of his " Historia 
rariorum Stirpium," 1576 (lib. i. c. 2), published five years 
after his journey in Spain. 

PassiJ/ora. — This genus is wholly omitted by De Candolle ; 
unaccountably so, considering how much Granadillas have 
been cultivated and prized in tropical countries. A note on 
the subject may not be out of place, as a species was culti- 
vated by our own Indians. 

•which M. De Candolle refers — the " Historia Stirpium," attributed to 
J. Bauhin, hut published long after his death (in 1551), with large additions 
by his son-in-law Cherler, and by Chabrseus and Graffenreid. Guillandi- 
nus, of Padua, in a treatise " De Papyro," 1572, named the " Tumatle 
Americanorum " as a species of " Pomum Amoris or Solanum pomiferum " ; 
and earlier, Matthioli had described it (Comment, in Dioscor., ed. 1559, 
p. 537J as a " kind of Mala insana " [Solanum Melongena], which was 
" beginning to be imported " into Italy, and which was " popularly called 
Pomi d'oro, that is, Mala aurea." 

Anguillara may have confounded the Tomato with another of the 
American Solanacea, introduced at about the same period — the Thorn 
Apple {Datura Stramonium) which Guillandinus (1572) named Mala 
Peruviana, and the French called "Pomme de Perou." — J. H. T. 



ORIGIN OF CULTIVATED PLANTS. 343 

As to early history and aboriginal nomenclature, Monardes 
(De Simplicibus Medic., c. 6Q~) says that the Granadilla was 
spontaneous in Peru, and that the fruit was highly esteemed 
by the Indians and by the Spaniards. Cieca de Leon (Chron. 
del Peru, c. 28), about 1550, saw it growing in the valley of 
the Rio Lile, near Cali (now in Colombia), and in the coun- 
try about Pasto. The flower and fruit are mentioned by J. 
de Acosta (Nat. and Moral Hist, of the Indies, b. ii. c. 38) 
— "the fruit sweet, and too sweet, in the opinion of some."' 
Lery (1557-8) does not appear to have found it in Brazil, 
but it was common there before the middle of the 17th cen- 
tury. Piso and Marcgrav (Hist. Nat. Brasil., 1648, pp. 70, 
106) reckon nine species or more, of which four were culti- 
vated — two especially for their fruits. The Tupi (Brazilian) 
name was " Mburucuia " (" Montoya," 1639), which Piso and 
Marcgrav, I. c, wrote " Murbcuia " ; the species which was 
generally cultivated for its fruit was " Mburucuia-guacu " 
(£. e., great " Murucuya ,r ). Father R. Breton (Diet. Ca- 
raibe, 16G5) gives " Merecoya " as the Carib name of the 
fruit -, but this seems to have been adopted from the Tupi — 
for in that language " Mburucuia " denotes the " fruit of a 
vine." 

It is remarkable that the Tupi (and Carib) name went with 
the plant and its fruit to the country of the Algonkins before 
the coming of Europeans. One species (Passiflora incar- 
iiitt(i) was cultivated by the Indians of Virginia. "They 
plant also Maracocks, a wild fruit like a lemon, which also 
increase infinitely " (Capt. John Smith, Gen. Hist., p. 29) : 
and, again (p. 25), Smith mentions the " fruit which the in- 
habitants call ' Maracocks ' — pleasant, wholesome fruit " — 
among " things which are naturally in Virginia." Strachey 
(Travaile into Virginia, 72, 119) describes the " fruit called 
by the natives a ' maracock,' which the Indians plant," etc. 

Although no longer planted, the fruit of the spontaneous 
plant is still eaten in the southern Atlantic States ; and its 
popular name, " May-pop," is probably the last stage of the 
Tupi original. 

Now our Passijfora incarnata is so like P. edulis (well 



344 REVIEWS. 

known in cultivation), the home of which is in Brazil, 1 that 
botanists have been unable clearly to distinguish the two, ex- 
cept by the fact that ours, dying down to the ground at ap- 
proach of winter, remains herbaceous. It occurs in a rather 
narrow geographical range ; and Dr. Masters, in his elaborate 
study of the order (Trans. Linn. Soc, xxvii. 641 ; see also 
Flora Brasiliensis), says that "being so far separate from 
the remainder of its allies of the same subgenus, [it] may be 
considered as an outlier." Altogether we may infer that the 
fruit and the name were originally derived from the same 
South American source. 

JLisa, Banana. — The author concludes, as did Robert 
Brown, that Banana and Plantain are varieties of one species ; 
also that this species is of the Old World ; that in all prob- 
ability it was not known in the West Indies when discovered 
by Columbus ; but that in respect to the western side of South 
America there is some evidence which is not easily ruled out, 
especially the statement of Garcilasso that the Peruvians had 
the Banana before the conquest, and of Stevenson ( Trav. in 
S. Amer.), who is said to have seen in the ancient Peruvian 
tombs beds made of Banana leaves. This is discredited because 
the author found beans in the same tombs, " et que la feve est 
certainement de l'ancien monde." But if Stevenson wrote 
beans, without doubt he meant the seeds of Phaseolus, not of 
Faba. It would rather seem that the Banana, like the Sweet 
Potato and Cocoa-nut, had early been transported over the 
Pacific. 

Phuscoliix vulgaris. Kidney Bean. — Three weeks after his 
first landing in the New World Columbus saw, near Nuevitas 
in Cuba, fields planted with " faxones and fabas very differ- 
ent from those of Spain," and two days afterwards, following 
the north coast of Cuba, he again found " land well cultivated 
with these fexoes and habas much unlike ours." " Faxones " 

1 Not Mexico, although indeed said to have been brought from New 
Spain to the garden of the Farnese palace, in Rome, as early as 1619. 
It is described, under the name of " Maracot " — with excellent plates 
showing the plant, the flower, and the fruit — by Tobias Aldinus, in 
" Earior. Plant. Horti Farnesiani " (lionise, 1625), pp. 49-59. 



ORIGIN OF CULTIVATED PLANTS. 345 

or " f sxoes " were — as Navarrete notes, " Colec." i. 200, 
203 — "the same as frejoles or judias," Spanish names for 
Kidney beans, which the Portuguese call " Feijaos." Oviedo 
(1525-35) speaks of the "fesoles, as the Spaniards call them, 
of which there are many kinds in the [AVest] Indias." These 
fesoles, he says (lib. vii. c. 18), " are called by Pliny ' fagi- 
voles ' : in Aragon we call them ' judias,' and the seeds of 
those of Spain and of this country are properly the same." 
The natives of Hispaniola raise these fesoles, but they are 
much more abundant on the main land, especially in New 
Spain and Nicaragua. " I have, in the province of Nagrando 
in Nicaragua, seen them gather a hundred hanegas (bushels, 
nearly) of these fesoles : and they also, in that country and 
other parts of that coast, have many other kinds of fesoles, 
besides the common sort : some have yellow seeds, others 
spotted," etc. In another place (lib. xi. c. 1), Oviedo, men- 
tioning plants that had been brought from Spain to Hispan- 
iola and other parts of America, " in the beginning," names 
" Fesoles, called in Aragon ' Judias,' and in my country [Cas- 
tile] ' Arvejas luengas ' : " but " of these, there is no need of 
bringing more seed, for in this island and on the main many 
bushels are harvested every year, and in the province of Ni- 
caragua they are indigenous (naturales de la misma tierra), 
and a great number of bushels are produced yearly of these 
and of other fesoles of other sorts and different colors," etc. 

From this time (1535) onward, neaily every writer who 
mentioned plants cultivated by the Indians named, together 
or in close connection, maize, beans, and pumpkins. Refer- 
ence to several of these writers has been made in our notes on 
Cucurbitce. Cabe^a de Vaca found beans cultivated by the 
Indians of Florida in 1528, and again near the western limit 
of his wanderings (in New Mexico or Sonora) in 1535. De 
Soto, at his landing in 1539, found " fields of maize, beans, 
and pumpkins," near Tampa Bay ; and at Coligoa (west of 
the Mississippi) " beans and pumpkins were in great plenty : 
both were larger aud better than those of Spain ; " and so, at 
other places, on his travels to the west and north. 

Jacques Cartier, the discoverer of the St. Lawrence, on his 



346 REVIEWS. 

first voyage, 1534, found that the Indians near the mouth of 
that river on the Bay of Gaspe had abundance of maize, and 
had " beans (febues) which they name ' Sahu,' " or (as spelled 
in the vocabulary printed with his Discourse du Voyage) 
" Sahe." 1 The " Bref llecit " of his second voyage, 1535-36, 
mentions the use of corn and beans by the Indians of the St. 
Lawrence — " bled & febues & poix, desquels ilz ont assez " 
(f. 24). 

Father Sagard in his " History of Canada " and in the ac- 
count of his journey to the country of the Hurons, 1625, men- 
tions the cultivation and use of " fezolles " by the Indians. 
The Hurons used in their succotash (" neintahouy ") " a third 
or a quarter part of their fezoles, called ogaressa " (Grand 
Voyage, 83, 138). 

Lescarbot, 1608, says that the Indians of Maine, like those 
of Virginia and Florida, plant their corn in hills, " and be- 
tween the kernels of corn they plant beans marked (feves 
riolees) with various colors, which are very delicate ; these, 
because they are not so high as the Corn, grow very well 
among it" (Hist. Nouv. France, ed. 1612, p. 835; see also 
p. 744). 

The relation of the voyage of Captains Amidas and Barlow 
to Virginia, 1584, mentions pease, melons, etc., at Roanoke 
Island, but does not name beans ; but Harriot, who accom- 
panied them on this voyage, includes both " Wickonzour, 
called by us pease," and " Okindjier, called by us beans," 
among the productions of that country. Capt. John Smith, 
who was in Virginia in 1607, and Strachey, who was there in 
1610, describe (in nearly the same words) the Indian manner 
of planting corn and beans : " they plant also pease they call 
assentamens, which are the same they call in Italy fagioli : 
their beans are the same the Turks call garvances, but these 
they much esteem for dainties " (Smith's Gen. Hist., 28 ; 
Strachey, Trav. in Virginia, 117). Evidently, these names 

1 The language spoken by these Indians was a dialect of the Huron- 
Iroquois group, and we trace the name sahe (as Cartier caught it) in the 
Mohawk osahe-ta, " fe'soles " of Bruyas (17th century), and the Onondaga 
ousahUa and hosah'eta, " poix, feve " (Shea's Onondaga Dictionary). 



ORIGIN OF CULTIVATED PLANTS. 347 

are confounded. Garvance was the French name of the 
Chick Pea (Cicer arietinuni), the Spanish garbanzo; and 
it is not probable that the Turks gave this name to any kind 
of beans ; while f agiuoli was the Italian equivalent of Latin 
phaseoli. Strachey's Virginian vocabulary gives assentamens 
(and otassentamens) for *' pease," and peccatoas, peketawes, 
for " beans." 

It must be remembered that at the beginning of the 17th 
century kidney beans — as well as vetchlings (Lathyrus) — 
were popularly regarded as a kind of pease, or "peason." 
Turner, in his " Names of Herbes," 1548, says that " Phasio- 
lus otherwyse called Dolichos, may be called in English ' long 
peasen or faselles ; ' . . . in French phaseoles : " and " Smi- 
lax hortensis, ... in French, as some wryte Phaseole . . . 
may be called in English Kydney beane," etc. (Eng. Dial. 
Soc, ed. 1881, pp. 62, 74). Lyte's " Dodoens," 1578, follows 
Turner for the English namets of Phaseolus, " Kidney beane 
and Sperage ; of some they are called Faselles, or Long Pea- 
son," etc. (p. 474) : his " common Peason " and " middle 
Peason" are Ervilia (JSrvwm Ermlia L?) and Pisum ar- 
vense L. ; while P. sativum is distinguished as " Great Pea- 
son, Garden Peason, and Branche Peason, because, as I 
thinke, they must be holpen or stayed up with branches " 
(id. 476). ' 

So, on the continent, the Spanish names for fesoles was 
"arvejas luengas" (Oviedo), i. e., long vetches, and the gar- 
den pea (_P. sativum') was " un corto genero de Arvejas " 
(Calepin's Diction., ed. 1616), a short kind of vetches. The 
confusion of names is frequent in writers of the 16th and first 
half of the 17th centuries, in French as well as English and 
Spanish. " Magno sane labore " — as Tragus found — 
" Phasioli, Orobi, et Pisa, nee non Cicer arietina, maxime 
agrestia, secernuntur," (Stirp. Hist., 1552, p. 613). 

Champlain uses " poix " and " feves," interchangeably, as 
names of the American Phaseoli. In Breton's " Diction. 
Fr.-Caraibe " (1666) the Carib name, " calaoiiana," stands for 
"poisde Bresil" and "febue de Bresil"; " pois rouges, dit 
Anglois, mibipi ; " and " Pois, manconti " — this being, prob- 
ably, the (introduced) P. sativum. 



048 REVIEWS. 

Sagard (1624-5) says that the Hurons called the coarser 
part of their pounded maize — after the meal had been sifted 
from it — " Acointa, c'est a dire Pois (car ils lui donnent le 
mesme nom qu'a nos pois) ; " and in his " Dictionnaire 
Iluronne," he has "Pois, Acointa," " Fezolles, Ogaressa;" 
whence we infer that French pease [P. sativum] were 
already cultivated by or known to the Hurons. The Abnakis 
of western Maine, in the 17th century, called pease, " awen- 
nootsi-minar," i. e. " French (or foreign) seeds." Tanner, 
1830, gives as the Chippeway name of the " Wild Pea-vine " 
[Phaseolus diver sifolius ?] " Anishemin," i. e. " Indian (or 
native) seeds." In nearly all North American languages, the 
names for kidney-beans (Phaseoli) are of earlier formation 
than those for garden pease. The latter are usually formed 
on the former : e. g. Chahta, tobi, " bean " ; tobi hullo, [wild] 
"pea"; tobi Mhint uni, "garden pea " (Byington) ; Dakota, 
<>"innkha, " bean " ; o n mnicha hmiya n ya n \_i. e. " round bean "], 
" pea." 

Without multiplying citations — we may assume that the 
" pease " and " poix " which early voyagers found cultivated 
by the American Indian were species of Phaseolus — not 
Pisum. 

Five and twenty years before the settlement of Virginia, the 
Indians of Carolina and Florida had " fine citroiiilles and 
very good beans " (Lescarbot, Nouv. France, 778). Lawson, 
1700-1708, gives a more particular description of the south- 
ern beans cultivated by the Indians. " The Kidney-beans," he 
says, " were here before the English came, being very plenti- 
ful in Indian corn-fields." " The Bushel bean," a sponta- 
neous growth, very flat, white and mottled with a purple figure, 
was trained on poles [P. multiflorus ?] : " Indian Rounceval, 
or Miraculous Pulse, so called from their long pods and great 
increase ; they are very good, and so are the ' Bonavies, Cala- 
vancies [= Garvances ?], Nanticokes,' and abundance of 
other pulse, too tedious to mention, which we found the In- 
dians possessed of when we settled in America "(Voyage to 
Carolina, pp. 76, 77). 



ORIGIN OF CULTIVATED PLANTS. 349 

In the northern States, we find little difficulty in establish- 
ing the identity of Phaseolus vulgaris with the beans culti- 
vated by the Indians at the first coming of Europeans. These 
were from the first distinguished as " Indian beans," from 
the Garden beans ( Vicia Faba) introduced by the English. 
In 1609, Hudson, exploring the river which bears his name, 
saw at an Indian village — in the vicinity of Schodac and 
Castleton, Rensselaer County, N. Y. — "a great quantity of 
maize or Indian corn, and beans of the last year's growth" 
(Hudson's Journal, in De Laet, 1625, b. iii. ch. 10, and 
Juet's, in Purchas : N. Y. Hist. Soc. Coll., 2 Ser., i. 300, 
325). 

1631-42. The Indians of New Netherland " make use of 
French beans of different colors, which they plant among 
their Maize. . . . The Maize stalks serve, instead of the poles 
which we use in our Fatherland, for the beans to grow upon " 
(De Vries, Voyages, transl. in 2 N. Y. Hist. Soc, iii. 107). 

1653. Van der Donck, in his " Description of the New 
Netherlands," distinguishes the beans cultivated by the In- 
dians before the coming of the Dutch, and the Turkish beans 
which had been introduced : " Of Beans there are several 
kinds ; but the large Windsor bean [ Vicia Fabti] . . . and 
the Horse-bean will not fill out their pods. . . . The Turkish 
beans which our people have introduced there grow wonder- 
fully. . . . Before the arrival of the Netherlanders [1614] 
the Indians raised beans of various kinds and colours, but gen- 
erally too coarse to be eaten green or to be pickled, except 
the blue sort, which are abundant," etc. He then describes 
the Indian mode of planting beans with maize, ut supra 
(N. Y. Hist. Soc. Coll., 2 Ser., i. 188-9). This is the only 
reference we have found to the introduction of any species of 
Phaseolus into North America. Van der Donck's book was 
written more than forty years after Hudson's coming, and the 
author first arrived in New Netherland in 1642. His state- 
ment as to the introduction by the Dutch of the best kind of 
" Turkish beans " for " snaps," salad, or pickling, is not to 
be accepted without reserve ; but the fact that Turkish beans 
" grow wonderfully, fill out remarkably well, and are much 



350 REVIEWS. 

cultivated," while the imported Windsor beans \_Vicia Faba] 
and horse-beans proved failures, is to be noted. 

Wood, who was in Massachusetts from 1629 to 1633, says 
that the Indians "in winter-time have all manner of fowles, 
Indian beanes, and clams " (N. E. Prospect, pt. 2, ch. 6). 
Roger Williams, 1643, gives the Indian name of these beans 
in the Narragansett dialect : " Manusqussed-ash " (plural) ; 
Cotton's Massachusetts vocabulary (1727-8) has (sing.) 
" Monasquisset, an Indian bean ; " President Stiles, about 
1760, heard the name in the Pequot dialect as "Mushquis- 
sedes" (MS. Vocab.) ; Zeisberger, 1776 and 1803, wrote it 
in the Delaware, with dialectic modification, " Malachxit ; " 
and we can trace it in the modern Shyenne " Monisk " (Hay- 
den's Vocab., 1862) and " Monchka." In the Chippeway, the 
kidney-bean has received — probably from some local variety 

— a different name: " Miskodissimin," i. c. "red-dyed seed 
(or fruit) ; " and this name, modified as " M'skochl-tha," was 
used by the Shawanees of Ohio. 

To return to New England, Josselyn, who was in this coun- 
try, 1638-9, and again, 1663-71, in his catalogue of " plants 
proper to the country," names " Indian beans, falsely called 
French beans :" "the herbalists call them kidney-beans, from 
their shape and effects. . . . They are variegated much [in 
size and color] ; besides your Bonivis and Calavances, and the 
kidney-bean that is proper to Roanoke : but these are brought 
into the country : the others are natural to the climate " (N. 
E. Rarities, p. 56 ; Voyages, p. 73-4). Here is reference to 
at least two species of American beans, one " proper to New 
England," the other from Roanoke — perhaps P. multiflorus. 

Besides the names already mentioned — "Monasquisset," 
with its variants — there is another, in northern Algonkin lan- 
guages, for kidney-beans, which must have originally belonged 
to some high-twining variety. Eliot used it, in the plural, for 
"beans" in 2 Samuel, xvii. 28, tuppuhquam-ash — which lit- 
erally signifies " twiners ;" and Rasles (1691-1700) gave, in 
the " Kennebec- Abnaki" of Maine, for "faseole," aHeba'kwe 

— from the same root. A modern Abnaki vocabulary shows 
that this name is still in use — as " ad-ba-kwa." 



ORIGIN OF CULTIVATED PLANTS. 351 

As to the American origin of P. lunatus, the Lima bean 
and its varieties, there seems now to be no question. " It is 
evidently," says M. De Candolle (p. 276), " a Brazilian spe- 
cies, dispersed by cultivation, and perhaps long ago natural- 
ized, here and there, in tropical America." 

But as to the origin of P. vulgaris he is not free from 
doubt. He finds " (1) that this species was not cultivated in 
ancient times in the East Indies, the southwest of Asia, or 
Egypt ; (2) that we are not absolutely certain that it was 
known in Europe before the discovery of America ; (3) that 
at that epoch the number of varieties suddenly increased in 
the gardens of Europe, and all authors began to speak of it ; 
(4) that the majority of species of this genus are found in 
South America ; and (5) that seeds which apparently belong 
to this species have been found in Peruvian tombs [at Ancon] 
of a date somewhat uncertain, mixed with many other species, 
all of which are American." (p. 275.) 

The proof that P. vulgaris (and P. nanus'), in varieties 
almost innumerable, were cultivated by the natives of Amer- 
ica before the coming of Europeans, seems to be conclusive. 
The resolution of M. De Candolle's doubts as to the American 
origin of the species must depend chiefly on the identification 
of the species known as Phaseoli (Phaseli, Fagiuoli, Feso- 
les, etc.), in Europe, before the discovery of America. This 
identification may not be impossible, but the space at our 
disposal will not permit us to attempt it in this article, or even 
to reexamine the authorities on which M. De Candolle admits 
the probability " that the Dolichos of Theophrastus was our 
pole bean (Jiaricot a rames), and the Fasiolos our cultivated 
bush bean (Jiaricot nain)" p. 271. At present, we have 
only to offer one or two notes. 

1. The distinction indicated by Galen (De Alimentis, lib. i. 
cc. 25, 28) between the Phasiolos (^ao-ioAo?) of Dioscorides 
and Phaselus (<^>ao-r;A.os) — presumably the " vilis faselus " of 
Virgil — if well founded, seems to have been lost sight of in 
the middle ages. In Italy, the Greek and Latin names 
Phasiolos, Faseolus, Faselus, Fasillus, etc., passed into the 
modern Fagiuoli. Piero de' Crescenzi, of Bologna, whose 



352 HE VIEWS. 

treatise on agriculture was written near the beginning of the 
14th century, in Latin, and translated into Italian about 
1350, mentions, among field plants, Faseoli (Fagiuoli), as 
well known ; " some of them are red, some white. . . . They 
arc planted conveniently among panick, millet, and chick 
pease ; they are also planted in gardens, among cabbages and 
onions.*' l It is not certain that the red and the white were 
of the same species, or genus, or that either was a species of 
Phaseolus, L. In the first half of the sixteenth century the 
white Phaseoli were the more common and less esteemed. 
The young and tender pods were eaten, with the included 
seeds, in salads, or boiled with other vegetables. 

Two other early figures show that the Faseoli were not so 
u \vell known " to the herbalists of the 15th and beginning of 
the 16th century as to Crescenzi in the 14th. One is from 
a Venice edition of the " Hortus Sanitatis," 1511 ; the other 
from the "Tacuini Sanitatis" of Elluchasem Elimithar, Stras- 
burg, 1531, p. 49. They are equally unlike the modern 
Phaseolus, the earlier figure in Crescenzi, and each other. 
The second may have originally been intended for a Teasel 
(Dipsacus sylvestris), the Virga pastoris of the herbalists. 
Calepin's Dictionary (ed. 1G16) says, s. v. Faseolns, that the 
name Fasilli is now given by the common people to " a spe- 
cies of Cicercula." This "vulgaris Phaseolus" of Matthioli 2 
and other Italian botanists of the period is figured and de- 
scribed in the later editions of his commentary on Dioscorides, 3 

1 " Be Agricultural lib. iii. c. 10 (Italian, Ed. Venice, 1504). The Latin 
text was first printed at Strasburg in 1471, and with figures, 1486 ; the 
Italian version was printed at Florence, 1478. The figure of Faseolus in 
the earliest (Latin) edition we have seen, without date, but probably of 
Louvain, about 1480, has little resemblance to the Phaseolus of modern 
botany. 

M. De Candolle remarks (p. 272) that "authors of the loth century 
say nothing of Faseolus, or any analogous name," and that " this is the 
case with P. Crescenzi," — referring to a French translation of Crescenzi, 
printed in 1539, which we have not seen. 

2 " Vulgares Phaseoli, quibus passim in cibis vescimur, dum satis in cam- 
pis virent, non repant," etc., Matth. " Apologia adv. Amathum," 1559, 
p. 33; " Vulgaris usns Phaseolus." Ibid. 31. 

3 Ed. C. Bauhin, 1598, p. 341. In the earlier edition (Venice, 1559, p. 



ORIGIN OF CULTIVATED PLAXTS. 353 

as having an erect stem, ternate leaves, white flowers, kidney- 
shaped seeds, white, "except the umbilicus, which grows 
black." It may be worth noting that " a black spot in the 
place of the cotyledon " was a characteristic of the Faseoli 
described by Albertus Magnus (13th cent.), which appears 
to M. De Candolle " to be the dwarf Haricot {P. nanus} of 
our epoch." This black spot is more strongly suggestive of 
Dolichos than of any known variety of P. vulgaris: e. g. 
Dolichos unguiculatus, L. (French, D. Mongettc, Banette, 
Haricot cornille^) — not mentioned by M. De Candolle, but 
much cultivated in Italy, and of which there are a great num- 
ber of varieties — which has seeds " marked by a prominent 
black spot about the umbilicus." * 

2. M. De Candolle (p. 272), with a reference to Delile and 
to Piddington's " Index," remarks that though " no Hebrew 
name corresponding to the Dolichos or Phaseolus of the bot- 
anists " is known, yet " a name less ancient, because it is 
Arabic, namely ' Loubia,' is found in Egypt for the Dolichos 
Lubia; and in Hindustani, under the form 'Loba,' for Pha- 
seolus vulgaris." This name seems to be clearly refei-able to 
the Greek. It has not been traced earlier than to Jahia ebn 
Serapion — an Arabian physician of the 9th or 10th century 
— whose work "De Simplicibus," compiled chiefly f rom Dios- 
corides and Galen, was translated into Latin in the loth cen- 
tury. 2 In a chapter (lxxxi) on " Lubia," i. e. " Faseoli," he 
quotes from Dioscorides the description of Smilax hortensis 
(j0T?aLa <rixL\a£) " whose seeds some call Lobia ; " and it is 
evident that the name Lubia (as it was transliterated from 
the Arabic text by the translator) was transferred to the 

2G4), Phaseolus is figured as a low, bushy, and spreading, but not twining 
plant. 

1 Several other species of Dolichos (e. g., D. sesquipedalis L., Ital. " Fag- 
iuolo Sparagio," Engl. " Asparagus bean ") are similarly marked. Vil- 
morin-Andrieux et Cie., " Les Plantes Potageres," 1883, p. 280. Other 
names for this species are : Germ. " Ostindische Jtiesel-Spargel Bohne," 
" Xagelische Fasel ; " Ital. " Fagiuolo dell' occhio ; " Span. " Garrubia," 
"Moncheta," " Judia de Careta." 

2 Milan, 1473, and Venice, 1479 ; but better known to botanists of the 
16th century in the Strasburg edition of 1531, edited by Otho Brunfels. 



354 REVIEWS. 

Arabic from the Greek of Dioscorides. It is probable, to say 
the least, that it has been rightly appropriated to Dolichos 
Lubia, Forskal (De Candolle, 278), rather than to any species 
of Phaseolus. 

The length to which our annotations have extended forbids 
all notice of the third part of this book. This, however, is 
very brief. It contains a tabulation of the plants of cultiva- 
tion, and of the results of the preceding discussion of them ; 
also an article on the regions in which the principal species 
have originated or have been brought iuto cultivation, in which 
it is stated that of the 247 species under investigation the Old 
World has furnished 199, and America 45, leaving three 
which are doubtful in this regard ; and the extreme poverty 
of the southern hemisphere beyond the tropics is a striking 
feature. An article on the number and nature of species 
cultivated at different periods is noteworthy. So, also, is the 
enumeration of the cultivated plants which are unknown in 
a wild state ; from which it is gathered that 27 species have 
never been found wild by any botanist, 27 more are doubtful 
in this respect, while 193 are of recognizable origin. 

Of the " Reflexions diverses," at the close, we note only the 
final one, that " In the history of cultivated plants I have 
found no indication of communications between the inhabitants 
of the Old and New World anterior to the discovery of Amer- 
ica by Columbus. The Scandinavians, who had carried their 
expeditions to the northern United States, and the Basques of 
the Middle Ages, who had extended their whaling voyages 
perhaps to America, would appear not to have transported 
a single cultivated species. The Gulf-stream has equally been 
without effect. Between America and Asia two transporta- 
tions may have been effected, one by man (the Batatas), the 
other either by man or by the sea (Cocoa-nut)." 

Perhaps the Banana should be ranked with the Sweet 
Potato in this regard. And we may merely conjecture that 
the Purslain came to our eastern coast with the Scandinavians 
or the Basques. 



GENERA PLAN T ARUM. 355 



BENTHAM AND HOOKER'S GENERA PLANTARUM. 

The completion of this great work l marks an era in sys- 
tematic botany. As the senior author is nearly eighty-three 
years of age, we may say that, essentially, it brings to a 
worthy close the long course of scientific labors of the most 
accomplished, sagacious, and continuously industrious phy- 
tologist of our day, almost the only survivor of those who 
personally knew Antoine Laurent Jussieu, and were asso- 
ciated with the elder De Candolle ; one whose line of author- 
ship, begun in the year 1826 (antedated only by that of the 
venerable Roeper), comes down to the present year, ap- 
parently, in almost unabated vigor, and is as remarkable for 
sustained importance as it is unparalleled in length. Upon 
the junior author, who should not yet feel the burden of years, 
however weighted with official cares, and upon such specialists 
as can be mustered, may devolve the responsibility of such 
addimenta as may be needed to bring the earlier volumes up 
to the date of the last, and even to supplement the work from 
time to time, as the " Genera Plantarum " of Linnaeus was 
supplemented by successive editors. But it is not probable 
that the task which has now been happily accomplished will 
be undertaken again, still less executed, before the twentieth 
century is well entered upon. Useful compilations we may 
expect, and monographs which may here and there better re- 
present advancing knowledge of particular groups ; but a 
production like this, covering the whole field of phaenogamous 
botany, ordinally and generically, and with the uniformity of 
treatment and scale thus secured by two close associates in 
one continuous work, cannot be looked for again for a long 
while. 

It has few predecessors. The first " Genera Plantarum " 
in fact (though not exactly in name) was that of Tournefort, 
published in the year 1700. Then and there, as Linnaeus de- 

1 Genera Plantarum ad Exemplariq imprimis in herbariis Kewensibus 
servata definita. Auctoribus G. Bentbam et J. D. Hooker. Loudiui, 1862- 
1883. (The Nation, No. 942, July 19, 18S3.) 



356 RE VIE WS. 

clared, genera of plants, in the sense of scientific botany, were 
first established. There; would probably be more recognition 
of this dictum if the present work were to be planned anew, 
and the genera which Linnaeus himself admitted as of Tourne- 
fort, along with those which modern botanists have restored, 
could have been attributed to this real founder, without 
thereby compromising the proper position to hold in respect 
to herbalistic and ancient names. The second " Genera 
Plantarum " was that of Linnaeus, in 1737, of which the 
last edition revised by the author himself was that of 17G7. 
The third was that of Jussieu, " secundum ordines naturales 
disposita," which appeared in the year 1789. That of End- 
licher — a monument of literary or bibliographical erudition 
rather than of botanical research — was brought out in the 
main between 1836 and 1843, at about the same time with 
the more unpretending synoptical compilation of Meisner. 
These were important in their way. But the " Genera Plan- 
tarum " of Bentham and Hooker, which began to be issued 
in the year 1862 and was finished in the spring of the present 
year, is the lineal successor of the three classical works above 
mentioned, that of Tournefort representing the botany of the 
close of the seventeenth century ; that of Linnaeus the first 
half, and that of Jussieu the latter part, of the eighteenth 
century. The present work — increased from the one small 
octavo of Linnaeus to three thick imperial octavo volumes of 
nearly 1200 pages each — stands in like relation to the nine- 
teenth century, and is based, like them, or even more than 
they, upon actual investigation, and upon the comparison of 
a vastly greater number of original types than was formerly 
possible. 

Unlike its predecessors, however, — and in this respect 
agreeing with the other great botanical work of the century, 
the " Prodromus " of De Candolle, — the whole of crypto- 
gamic botany is omitted. This vast field must be left to 
specialists. Fries, of Upsala, who died a few years ago at 
a ripe old age, was the last phaenogamic botanist who was at 
the same time master of one or two cryptogamic orders ; and 
now even the best of cryptogamists can hardly aspire to more 



GENERA PLANT ARUM. 357 

than a general and superficial acquaintance with any other 
department than the one to which he devotes himself. This 
inevitable state of things has its disadvantages. The reasons 
for it do not really apply to the Ferns and their allies, and it 
was naturally expected, as it is much to be desired, that these 
should enter into the present work. May we hope that this 
still may be ? 

Some idea of the progressive enlargement of the field may 
be had by a comparison of the number of genera characterized 
in these successive works. The phaenogamous genera of 

Linnaeus, " Gen. PI.," ed. 1, a. d. 1737, were . . 887 

" ed. 6, a. d. 17G1, "... 1,189 

Jussieu, " . . . . a. d. 1789, "... 1,707 

Endlicher, " . . . . A. r>. 1813, " (about) 6,400 

Bentham&> K a. d. 1883, "... 7,585 

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 95,620. In round 
numbers, it may fairly be said that about 100,000 species of 
phamogamous 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?, Leguminosce, Orchidece, Ru- 
biacece, Graminem. The high standing of the Orchid family 
in the list will be a surprise to many. Linnasus knew only a 
hundred species ; five thousand is now a moderate estimate 
— about half as many as there are of Composite^, 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 



858 HE VIEWS. 

over Orchids is owing to the 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. 1 

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 commentary, was pub- 
lished by De Candolle early in the autumn of the same year ; 
and an English translation, made by the lamented Dr. TTed- 
dell, appeared early in 1868. The " Laws," but without 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 application in leading 
cases ; but many practical questions, as well as conflicts of 

1 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. No small 
discussion upon certain details 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 summed up 
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 addition? are printed in italic type, so 
that they may be seen at a glance. 

Vie 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 pluenogamous 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 



300 REVIEWS. 

discussing the cases which still need adjustment, some of them 
of considerable consequence in phytographv. 

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, whose 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 between 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 principle of nomenclature, and as prevailing 
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 making 
old and accepted names better, more classical, or more signifi- 
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 up 
in the author's order the points which we wish to specify or 
to comment on. 

Article 6 of the code declares that " scientific names should 
be in Latin. When taken from another language, 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 Linnaeus should be the model. 
It is the classical language of botany, and is much more pre- 
cise than the Latin of antiquity, in which 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 6G, as we shall see. But, in fact, even technical 
lamruao-e cannot alwavs avoid ambiguities and the use of 
words in senses which have to be determined by the context. 
Folium^ for example, in botanical description, 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. 



QAO 



b'J, HE VIEWS. 

" 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 bls , 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 species 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 with 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. 1 

De Candolle's main remarks upon article 15 relate to the 

1 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 ; ouly 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. 1 The counterpart, Phauerogamia (or Phaenogamia), 
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 Candolle'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, Lappa major, 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. 



3G4 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 Linnaeus 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 Linnaeus ascribes, but that " he kept a good 
many adjective names for genera (Acetosa, Bermudiana, 
etc.)." Since Linnaeus 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, Linnaeus 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 Linnaeus is not to be corrected ; so 
Gloriosa, Impatiens, etc., remain. 

Are we equally to follow Linnaeus 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., Corisperraum, A. Juss., Olea, Tourn., 
Justicia. Houst., Dianthera, Gronov., Lycopus, Tourn., Lin- 
naea, 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 Tournefortian genera suppressed 
by Linnaeus 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 Linnaeus 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 " Linnaea, 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 Plantaruni," against which a protest was 
made in this Journal when the first 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 Gaertner 
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 Linnaea, 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 Linnaeus 
from Tournefort, etc., and so accredited, should continue to 
be thus cited ; that the date 1737 (Linnaeus, " 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. 1 Since 
Sa.cifrugacece, Myrsineacece, Styracacece, Ge?itianacece, Nyc- 
tayinacece, and even Lauracece and Juglandacem were adopted 

1 Cruciferce, Leguminosce, Umbelliferce, Compositce, Labiates, and the like, 
ai'e 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 " Prodromus," it seems to us retrograde and unadvis- 
able to have gone back in the new " Genera Plantarum " to 
SaxifrtKjav, and the like ; and this upon no obvious prin- 
ciple, as we have Samydacece, Cornacece, etc., brought into 
harmony with the rule. And if in the " Prodromus " we 
have JStyracacece, why not also Salicacem f And if, in the 
" Genera Plantarum," Styrax could take the acece termina- 
tion as Styracece, 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 : Namely, that too rigid adherence to the rule that 
names of suborders, tribes, etc., shall end in -e.ee, 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 — Saurai/jeac, 
Spiraeeae, Catesbaeeae, Jaumeeae, Thymleecae, and Moraeeae 
are the worst instances, and would justify any infraction of 
rules. 1 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 Iridece. 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 -q by e is not in full accordance 
with Latin usage, as witness bibliotheca, dialectica, Jlecubea, 
etc. 

1 Far better to write Spirceacece, with De Candolle. 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 acecc, such as 
Vernoniacece and Eupatoriacece, adopted b) 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 
Trichocladus and Ancistroeladus would have been euphonious 
as Trichocladia and Ancistrocladia, and very little longer. 
Acanthopanax, Didymopanax, Dimorphochltimys, Trigono- 
chlamys, Aulacoealyx, Pellacalyx, Microcharis, and the 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, 
Leueas, 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 
Plantarurn," — 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- 
guinaria, 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 RE VIE WS. 

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., Clusii) when the person com- 
memorated was a discoverer, describer, or an illustrator of 
the species, but were in adjective form (e. g., 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 nemorum, Con- 
volvulus sepium, Rhus vemix, Chamcerops hystrix (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 Scep>- 
trum ; " Ardisia Pickeringia, a species of this genus which, 
mistaken by Nuttall for a new one, had been named Picker- 
ingia ; Rudbcchia Heliopsidis, a Rudbeckia facie Heliopsi- 
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 
BJwt 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 — happily as a 
synonym — Heleniwn /Semincwiense, published by a Pro- 
fessor who thought he had discovered a new species of Hele- 
niuui 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 zona/e, 
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 specimens 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 might 

1 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. ^Ve 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 hecause the groups are 
new, hut because they are old and entirely familiar. Nobody will ever 
suppose that Ranunculacece, Clematidcce, 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 Plantarurn 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 Plantarurn," 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 bulbosus 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 plant, 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 " Mathi- 
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 " Mathiola tristis, Linn.," we make an untrue 
statement. Linnaeus had a wholly different genus Mathiola, 
and no J\L 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 synonym}' 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. (sub-Cheiraiitho)," our longer phrase still wants 

stated without somewhat 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 
" Mathiola tristis, Linn. (Cheiranthus, 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 '•'■Mathiola tristis, Brown (Cheircmthus 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. Linnaeus 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-Linnaean 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 para- 
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, Martius, etc., followed this course. It is then a useless 
complication of many modern naturalists to append ' mihi,' 
' nobis,' ' sp. nov.,' ' gen. nov.,' etc., to a new name. A large 
majority of species, genera, and families were 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 author 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 
worded, " 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 which 
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 Linnaean limits or signification, and the new rule 
was practically proved to be a necessity. 

Anions: 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 tins, 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 would 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 UmbelliferoR 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 dealing with such names 
as the above mentioned after they have been published, L 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 Candolle gives us " Eu- 
lophus, Nutt.," as the name of a genus communicated by 
Nuttall. with a specimen, for the purpose 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 
Linnaeus published as " Linnaea, authore Clariss. Dr. Grono- 
vio," we are to cite as Linnsea, Linn. This is not only quite 
contrary to the practice of botanists from Linnaeus down to 
De Candolle 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 ineditecl 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 
dliatum, 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 dliatum, 
Dougl.," 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." Xot so, however, with " Gillies 
in Hook." Sir AVilliam 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 



876 REVIEWS. 



general works, except in the bibliography or full reference ; 
when of course the " Leptocaulis iner)7iib, 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 
younger Richard and Kunth essayed to do. The same holds 
for the work of Solander in the first edition of Aiton'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 upon 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 abbi-eviated 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 Linmeus, "DC." for De Candolle, "H. B. K." 



BOTANICAL NOMENCLATURE. 377 

for Humboldt, Bonpland, and Kunth. We are not sure that 
De Candolle would favor the latter. 

A series of remarks is made upon articles 29-G6, 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 sagittifolia 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 Mintlie, Hydrocotyle brevipes to H. 
brevipedata, Cactus to Cactos, Arum to Ai*on, 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 
Linnaeus (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 
corrections 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 
should be written as near as may be with the original ortho- 
graphy of the person's name. "Astragalus aboriginorum" 
is neither a typographical nor a clerical error. It is a hard 
rule that forbids us to write " aboriginum," still retaining 
Richardson's name as authority. 

Botanists may take more kindly to the rule when applied 
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 Robert 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 grammatlcum, — 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 Linnseus and Lis contemporaries, notwithstand- 
ing classical faults, enables us to retain such familiar names 
as Ranunculus acris, Lathyrus pahistris 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 /cerrpov, and not 
the impossible word Kkvrpov, must have been intended by 
Borckkausen, we think it should now be maintained, although 
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 GO, 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, which 
was treated with some fullness in the " Phytographie." The 



380 REVIEWS. 

remark is here repeated that the 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 only, 
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. 381 

the 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 Candolle would call micromorphic botanists, like Jordan 
and Gandoger, who abandon the Linnaean 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. 
When 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 the 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 naming 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 Compositce nearly one half. For 
example, to Helenium he has referred Gaillardia, Actinella, 
Cephalophora, etc. ; to Tagetes he has referred Dysodia, 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 weight. 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 appears to be the only 
workable system. As already intimated, the right assignment 
of specific names in reconstructed genera requires particular 
knowledge 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. 



BALLS FLORA OF THE PERUVIAN ANDES. 

The personal observations and the collections upon which 
this essay 1 was founded were made in April, 1882, in an 
excursion by railway from Lima up to Chicla, which although 
only seventy-five miles in distance, is at the elevation of 12,220 
English feet. Much to bis 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 pha?nogams 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 Compositce are the 
Mutisiacece. Mr. Ball, referring to Bentham's indication of 
the complex affinities of this group, ventm^es " 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.) 



BALL'S FLORA OF THE PERUVIAN ANDES. 385 

of Compositce 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 Compositce 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 mo mtain 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 Compositce 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 Compositce. 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 Compositce and in such-like orders ; the other, 
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, reco<rnizini- 
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 rule 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 



BALLS FLORA OF THE PERUVIAN ANDES. 387 

the present flora of our temperate zone to a former high- 
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 MeIasto?nacece, 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 through 
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. xxvii., 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 Rocky 
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 JPolemoniacece, of which 
about 140 species belong to that region, are represented in 
the Andes by five species of Gilia, one of Collonia, and by 
the endemic genus Cantua. They have sent to the Old 
AVorld 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 Hydrophyllacece, 
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 Rimac in the Peruvian Andes, with various an- 
notations and the characters of some new species, concludes 
the present interesting contribution to Andean Botany. A\ e 
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. 



BALL'S FLORA OF THE PERUVIAN ANDES. 389 

Er odium cicutariinn. — 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 height 
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 than 
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 the 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 eircinata, which extends almost from one end to 
the other of the American continent, is said to be singularly 
constant, exhibiting no marked varieties. Bat 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. 



Acanthaceee 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. 
Ampelideoz, Planchon's Monograph of, 

250. 
Ampelopsis, characters of, discussed, 

251. 
Ancestry of plants, 2G9. 
Andes, Peruvian, Ball's Flora of, 3S4. 
Anemophilous plants, 234. 
Anthephora axilliflora, 113. 
Apple, effect of foreign pollen on an, 

described (note), 189. 
Aquilegia jlavescens, 185. 
Archasology, 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. 

Balanophorece, 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, 1S1. 

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 Euphorbiacea, 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, 2S4. 

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, 3S7. 



392 



INDEX. 



British Flora. Bentham's Hand-Book 

of. L04. 
British flora, statistics of, 105. 
Buckloe dactyloides, 113. 
Buffalo-grass, 112. 
Bu/bochcete, nature of, 92. 

Cala nthera dactyloides, 1 1 3. 

Cassava-plant, origin of, discussed, 
320. 

Cell-formation, 54. 

Cell-markings, 53. 

Cell, physiological condition of, 55. 

Cell, the vegetable, Von Mold's, 51. 

Ceropegia, revolution of growing stem 
of, 162, 1G7. 

Chamailirium Carolinanum, 281. 

Champlain, Les Voyages du Sieur de, 
S4, 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. 

Coba?a scandens, mode of growth of, 
177. 

Color in flowers, purpose of, described, 
227. 

Comandra, 29. 

Composite in the Prodromus, 16. 

Composites of the Peruvian Andes, 

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. 
Cucurbita, Naudin on the genus, 83. 
Cucurbita maxima, 83, 332. 
Cucurbita moschata, 83, 332. 
Cucurbita Pepo, 83, 332. 
Cucurbita verrucosa, 85. 
Cucurbitacecc, cross-breeding of, S6. 
Cucurbitaceoz, origin of cultivated 

species of, 329. 
Cucurbitaceoz, tendrils of, 174. 
Cultivated Plants, De Candolle's Origin 

of, 311. 
Cultivated plants, improvement of, 

109. 
Cultivated plants, origin of, 69. 
Cupuliferoz, 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, 15S. 
Darwin's Power of Movement in 

Plants, 304. 
Decaisne's Monograph of the Genus 

Pyrus, 186. 
De Candolle's Prodromus, 15. 
De Candolle's (A.) Ge*ographie Bota- 

nique, 67. 
De Candolle's (A.) Monograph of 

Smilaceaz, 249. 
De Candolle's (A.) New Monographs, 

248. 
De Candolle's (A.) Nomenclature, 358. 
De Candolle's (A.) Origin of Cultivated 

Plants, 311. 
De Candolle's (A.) Phytography, 282. 
De Candolle (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 



\J 



Diccntra, origin of name discussed, 

378. 
Diclytra, origin of name discussed, 

• i-o 

• J IO. 

Dielytra, origin of name discussed, 

378. 
Dioncea, insectivorous habits of, 207. 
Dioscorea saliva, origin of, discussed, 

322. 
Douglasia, 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. 

Endlieher's Genera Plantarum, 33. 

Engelmann's Notes on the Genus 
Yucca, 196. 

Engelmann on the Butfalo-grass, 112. 

Epiontology defined, 131. 

Epping Forest, 253. 

Erodium cicutarium, distribution of, 
389. 

Euphorbiacecc, Bentham on, 259. 

Fecundation, Process of, in the Vege- 
table Kingdom by Radlkofer, 91. 

Fertility, degrees of, in crossed seed- 
lings of Ipomaia purpurea, 221. 

Fertilization, self-, Henslow on, 263. 

Fertilization of flowers, relation of in- 
sects to, 2:)-"). 

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, 269. 
Fossil plants, nomenclature of, 3S0. 
Fruit, ripening of, 78. 
Fruits, nomenclature of, 78. 
Fucaceai, fecundation of, 93. 
Fucus, note on Decaisne and Thuret's 

Memoir of, 59. 
Fungi, note on LeVeille's paper on, 59. 

Gaura parvifiora, 267. 

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, Endlieher'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. 
G^ographie Botanique, A. De Can- 

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. 

Ilamamelis Virginica, properties of, 
79. 

Harvey's South African Plants, 36. 

Hastingsia, 281. 

Heer, appreciation of, 270. 

Helianihus argophyllus. experiments 
on growth of, 101. 

Helianihus giganteus, origin of Jerusa- 
lem Artichoke discussed, 315. 

Helianthus tuberosus, origin of, dis- 
cussed, 314. 



394 



ISDEX. 



Henfrey's Botany. 72. 

Henslow, George, Self-Fertilization of 

Plants, 263. 
Herbaria, their history. 004. 
Heredity and variability in plants, 

nature of, 212. 
Hesperoenide tenella, 90. 
lhttromeles, 195. 
Heterosmilax, 249. 
History of herbaria. .'Ii >4. 
llollisteria, 2S2. 

Hongkong, Bentharn's Flora of, 117. 
Hook-climbers, mode of growth of, 

171. 
Hooker on Balanophorece, 94. 
Hooker on distribution of Arctic 

plants, 122. 
Hooker on Welwitschia, 151. 
Hooker and Ball's Tour in Maroceo, 

255. 
Hooker and Thomson" s Indian Flora, 

62. 
Hop, revolutions of growing stem of, 

100. 
Hops, native in North America, 328. 
Hamulus Lupulus, native in North 

America, 328. 
Hybrids, characters of, 212. 
Hybridization of Lilies, 208. 
Hybridization, views upon, in Flora of 

India, 63. 
Hydrophyllacece in the Prodromus, 20. 

Imbrication, quincuncial, 120. 

Improvement of cultivated plants, 100. 

Indian Flora, Hooker and Thomson's, 
62. 

Initials, capital, use of, discussed. 48, 
381. 

Insectivorous Plants, Darwin's, 206. 

Insects in relation to the fertilization 
of flowers, 235. 

Ipomcea purpurata, Darwin's experi- 
ments in fertilizing, 218. 

Japan. Siebold's Flora of. 07. 

Japanese flora contrasted with North 
American, 38. 

Jerusalem Artichoke, origin of, dis- 
cussed, 014 



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, 280, 301. 

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. 

Liliacece, North American, 27S. 

Liliacece, Watson's arrangement of, 278. 

Lilies, hybridization of, 238. 

Lillium Grayi, 281. 

Lillium Parkmanni, 238. 

Lima Bean, origin of, 05 1 . 

Lindley's Natural System of Botany, 
1. 

Literary material, methods of collect- 
ing, discussed, 287. 

Longevity of trees, 82. 

Lophospermum scandens, mode of 
climbing, 170. 

Lycopersicum 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 utilissima, 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 sempcrfiorens, mode of 

growth of, 171. 
Mt liacea;. monograph of, 249. 
Mold's Vegetable Cell, 51. 
llonanthochloe, 114. 
Monographs of plants discussed, 290. 



INDEX. 



395 



Monstrosities, vegetable, 40. 
Moquin-Tandon'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, 308. 

Names, methods of abbreviation dis- 
cussed, 376. 

Names of natural orders, 11, 365. 

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, 36S. 

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. 

Nomenclator, Agassiz's Zoological, 41. 

Nomenclature, A. De Candolle's, 35S. 

Nomenclature, Bentham's views on, 
206. 

Nomenclature of fossil plants, 380. 

Nomenclature of organs, 294. 

North Carolina, trees of, 115. 

NyctaginacecB in the Prodromus, 28. 

Nyctotropism explained, 307. 

Oakesia sessilifolia, 280. 

( );iks, 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 p ants, fertilization of, 

267. 
Parkmau, F., on the Hybridization of 

Lilies, 238. 
Parrya macrocarpa, 185. 
Parthenocissus, 252. 
Passiflora gracilis, movements of, 173. 
Passiflora, history of cultivation of, 

342. 
Passiflora 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. 
Pkacelia circinata, 389. 
Phaseolus lunatus, origin of, 351. 
Pkaseolus vulgaris, history of, 344. 
Photinia. 195. 
Phyllotaxy, 76. 
Physiology of plants, 81. 
Phytogamy, 241. 

Phytography, A. De Candolle's, 282. 
Piclceringia, 15. 
Pilea purnila, 90. 

Pinguicula, glandular hairs of, 209. 
Pirus, 89, 329. 



396 



IXDEX. 



Pistillum, use of the term discussed, 
297. 

Planchon's Monograph of Ampelideie, 
250. 

Plant Archaeology, 200. 

Plants, ancestry of, 269. 

Plants, anemophilous, 234. 

Plants. Climbing', Darwin's Movements 
and Habits of, 158. 

Plants. Cultivated, A. De Candolle's 
Origin of, 311; improvement of, 
L09. 

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, 08. 

Plants, Self-fertilization of, by George 
Henslow, 203. 

Plants, self-sterile, 224. 

Poinsettia, adaptation to insect-fertil- 
ization, 204. 

Pollen, direct action of. on ovary and 
ovules, 87 ; value of, discussed by 
Darwin, 228. 

Popular names of British plants dis- 
cussed, 100. 

Portulaca oleracea, introduction into 
North America, 320. 

Potato, origin of, discussed, 69 note, 
317. 

Primula Sinensis, glandular hairs of 
the leaves of, 209. 

Primulacece in the Prodromus, 18. 

Prodromus. De Car.dolle's, 15. 

Proserpina. Ruskin's, 199. 

Pruinosus defined, 297. 

Publication of botanical works dis- 
cussed, 2 v ". 

Publicity of publication discussed, 
369. 

Piirslain. introduction into North 
America, 320. 

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. 
Qut reus Ilex, 139. 
Quercus Eobur, 138, 139. 
Quincuucial 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. 
Ranunculus alismcefolius, 184. 
Pa nunculus fascicularis, 184. 
Ranunculus orthorhynchus, 184. 
Red Clover, fertilization of, 226. 
Restiaceai, monograph of, 24S. 
Rhamnacece, stamens of, 76. 
Rltijjogonum, 249. 
Riddles, botanical, 291. 
Root, mode of penetration into the 

soil. 300. 
Root-hairs, 75. 

Root-tip, functions of, discussed, 307. 
: Rosa setigera, mode of growth of, 171. 
Ruskin's Proserpina, 199. 

Salsolacece in the Prodromus, 26. 
! Santalaceo; in the Prodromus, 28. 

Saporta's Monde des Plantes avant 
P Apparition de 1' Homme, 209. 

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 discussed, 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 Candolle's Monograph 

of, 249. 
Smilacece, distribution of, 250. 
Smilacina, nomenclature of, 261. 
Smilax, 249. 

Soil, soluble matters in, 103. 
Solarium tuberosum, origin of, discussed, 

317. 
Species, A. De Candolle'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, difficulties 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. 

Utricular ice, 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, 
2S7. 

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 Liliaceoz, 

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, Engelmami's notes on, 196. 
Yucca, anthesis of, 197. 
Yucca, nature of the fruit of, 198. 



diversity of Toronto 
Uhra'y 




Acme Library Card Pocket 
LOWE-MARTIN CO. LIMITED