I1I2 
NATURE 
[Fune 8, 1871 

thorough acquaintance with the questions to be attended to. It 
is true that the artist working independently and copying me- 
chanically may serve as a check on the naturalist, who in minute 
microscopic examinations may be apt to see too much in con- 
formity to preconceived theories ; but that is not often the case, 
the most satisfactory analytical drawings I have always found to 
be those made by the naturalist’s own hand, and I havelong felthow 
much my own inability to draw has detracted from the value of 
botanical papers I have published. And thirdly, when we consider 
that the great advantage of an illustration over a description is, 
that the one gives us ata glance the information which we can 
only obtain from the other by study, we require that each draw- 
ing or’plate should be as comprehensive as is consistent with 
clearness and precision. Outline drawings or portraits without 
structural details often omit the essential characters we are in 
search of ; where details are unaccompanied by a general outline, 
we miss a great means of fixing their bearing on our minds. 
Structural details may also equally err in being too numerous or 
too few, or too large or on too smalla scale. If the plate is 
crowded with details of little importance, or which may be 
readily taken from the general outline, they draw off the atten- 
tion from those which it is essential should be at once fixed on 
the mind, andif enlarged beyond what is necessary for clearness, 
they require so much the more effort to comprehend them, unless 
indeed they be destined to be hung up on the walls of a lecture- 
room. I believe it to be the case with some drawings of the 
muscles of vertebrata, or of the internal structure of insects, as 
I know it to be with those of ovules and other minute parts of 
flowers of the late Dr. Griffith and others, that with their very 
high scientific value, their practical utility is much interfered 
with by the large scale on which they aredrawn, A great deal 
depends also on the arranzement in the plate, always keeping 
in mind that the object is not to please the eye, but to convey 
at one view as much as possible of comparative information 
without producing confusion. 
Biological illustrations in general have much improved in our 
time. Itis true that some of the representations of animals and 
plants dating from the middle cf last century will enter into 
competition with any modern ones as to the general outlines 
and facies, but analytical details were almost universally 
neglected, and colouring when attempted was gaudy and 
unfaithful. At present I believe we excel in this country 
in the general artistic effect, as unfortunately also for the 
naturalist in the costliness, of our best zoclogical and bo- 
tanical plates; the French are remarkable for the selection, 
arrangement, and execution of the s-ientific details, and as a 
model I may refer to some of the publications of the Paris 
Museum, such as the Malpighiacee of Adrien de Jussieu, 
and also for the excellent woodcuts illustrating their general and 
popular works ; the Germans and some Northern states for the 
admirable neatness of microscopic and other minutize executed 
at a comparatively small cost, owing partially at least to the use 
of engravings on lithographic stone. 
4. Written Descriptions are what we must chiefly rely upon to 
convey to the general or to the practical naturalist the results of 
our studies of animals and plants ; but descriptions are of two 
kinds—individual descriptions and descriptions of species, 
genera, or other races. The former are like preserved specimens 
or delineations, materials for study, like them they require in 
their preparation little more than artistical skill guided by a 
general knowledge of the subject; but abstract descriptions, 
whether specific or relating to races of a higher degree, require 
that study of the mutual relations of individuals and races and 
their consequent classification which constitute the science of 
systematic biology, and this distinction should be constantly kept 
in view for the just appreciation of all descriptive works. Any 
tyro can with care write a long description of a specimen unim- 
peachable as to accuracy, but it requires a thorough knowledge 
of the subject anda keen appreciation of the bearings of the 
points noticed to prepare a good description of a species. For 
the latter to be serviceable it must be accurate, it must be full 
without redundancy, it must be concise without sacrificing clear- 
ness, it must be abstractive not individual, and lastly, the most 
difficult qualification of all and that which constitutes the main 
point of the science, the abstraction must be judicious and true 
to Nature. 
The paramount importance of accuracy is too evident to need 
dwelling upon. We are all liable to errors of observation. Im- 
perfect vision or instruments, optical deceptions, accidentally 
abnormal conditions of the specimen examined, hasty apprecia- 

tion of what we see from preconceived theories, are so many of 
the causes which have occasionally led into error the most 
eminent of naturalists, and require to be specially guarded against 
by repeated observation of different specimens and constant 
testing at every step by reasonings from analogy. Errors once 
established on apparently good authority are exceedingly difficult 
to correct, and have been the source of many a false theory. 
Where loose examination and hasty conclusion have been fre- 
quently detected, we can at once renounce all confidence in an 
author’s descriptions—in his genera and species—unless con- 
firmed from other sources, but an accidental oversight on the 
part of a naturalist of established reputation is the most difficult 
to remedy, notwithstanding the eagerness with which some 
beginners devote themselves to hunting them out. No botanist 
was, I believe, ever more careful in verifying his observations 
over and over again, and in submitting them to the tests supplied 
by the extraordinary methodising powers of his mind, than 
Robert Brown, no one has ever committed fewer of what we call 
blunders, or established his systematic theories on safer ground, 
yet even he has been detected in a few minor oversights, eagerly 
seized upon by a set of modern speculative botanists, lovers of 
paradoxes, as justifying them in devoting their time and energies 
to the disputal of several of his most important discoveries and 
conclusions. 
The value of a description as to fulness and conciseness is 
practical only, but in that point of view important. A descrip- 
tion, however accurate, is absolutely useless if the essential points 
are omitted, and very nearly so if those essential points are 
drowned in a sea of useless details; the difficulty is to ascertain 
what are the essential points; and hence one of the causes of the 
superiority of monographs and floras over isolated descriptions, 
such as those of Zoologies and Botanies of exploring expeditions, 
which I insisted on in my address of 1862; in the former the 
author must equally examine and classify all the allied races, 
and thus ascertain the essential points ; in the latter case he is too 
easily led to trust to what he believes to be essential. My own 
Jong experience in the using, as well as in the making. of botanical 
descriptions, has proved to me how difficult it is to prepare a 
really good one, how impossible it is to do it satisfactorily from a 
first observation of a single specimen. However carefully you 
may have noted every point that occurs to you, you will find that, 
after having comparatively examined other specimens and allied 
forms, you will have many an error to correct, many a blank to 
fill up, and much to eliminate. I have more than once had to 
verify the same species in two authors, the one giving you a 
character in a few lines which satisfies you at once, the other 
obliging you to labour through two or three quarto pages of minute 
details, from which some of the essential points are omitted. 
But the great problem to be solved at every stage in systematic 
or descriptive biology, and that which gives it so high a scientific 
importance, is the due detection and appreciation of affinities and 
mutual relations, and in this respect the science has made immense 
progress within my own recollection, and especially during the 
last few years the gradual supplanting of artificial by natural 
classifications has been too often commented upon to need repeti- 
tion. It is now, I believe, universally admitted that a species 
consists of individuals connected together by certain resem- 
blances or affinities the result of a common descent. It is 
also acknowledged that for scientific purposes these species 
should be arranged in groups according to resemblances or 
affinities more remote than in the case of species, although here 
commences the great difference of opinion as to the meaning of 
these remote affinities, whether they also are the result of a 
common descent, or of that supposed imitation of a type which 
I have above alluded to. For those, however, who have once 
connected affinity with consanguinity, it is difficult to re- 
cede from so ready an explanation of those mysterious 
resemblances and differences, the study of which must 
be the ruling principle to guide us in our  classi- 
fications. | All this has now been fully explained by more 
able pens than mine ; my only object in repeating it is to point 
out clearly the need of treating all systematic groups from the 
order down to the genus, species, or variety, as races of a similar 
nature, collections of individuals more nearly related to each 
other than to the individuals comprising any other race of the 
same grade, and of abolishing the use of the expression ¢yfe of 
a genus, or other group, in any other than a purely historical 
sense, as a question of nomenclature.* If a genus has to be 
* For the purposes of instruction some one species is often named as @ 
type of a genus, that is to say, as fairly representing the most prevalent 
