vi Supplement to ‘ Nature,” May 2, 1912. 
compendium of the details and literature of the 
subject, the English translation of which is in 
process of issue by the Oxford University Press 
in several volumes, there remains an obvious need 
for a more handy book dealing with the funda- 
mental principles of the subject from the modern 
point of view. Such a work would prove of 
greatest value both to the general reader and to 
those who have experienced a difficulty in obtain- 
ing a concise arrangement of the subject for teach- 
purposes. The volume issued by Prof. 
Kirchner has been designed to fulfil these require- 
ments, and the work has been carried out in a 
particularly effective and satisfactory manner. In 
a small volume of about 4oo pages, with large 
clear print and more than 160 detailed pen- 
sketches, many of them full-page blocks, a very 
comprehensive view of the whole question has 
been combined with a semi-popular mode of pre- 
sentation, which renders the material readily 
accessible to all. There are no footnotes and no 
direct references to literature, and the treatment 
of the subject is simplified as much as possible, 
since those requiring further detailed information 
with regard to particular plants will consult the 
volumes of Knuth. 
The book is divided into sixteen chapters, the 
subject-matter of which would be admirably 
adapted for a course of about twenty lectures, in- 
cluding a review of the entire range of phenomena 
associated with the reciprocal relationship of 
flowers and insects. An introductory chapter on 
the mechanism of pollination, and its relation to 
subsequent fertilisation, is succeeded by dis- 
cussion of the different methods of pollination and 
the essential difference between cross- and self- 
pollination, special emphasis being laid on the 
deduction that in special circumstances autogamy 
may be as essentially important to plant life as 
allogamy, and that it is impossible to understand 
the biology of the flower without taking into 
account the ecological relationships of the whole 
organism. The consideration of the special 
characteristics of insect-visited flowers involves 
the discussion of the origin and use of nectar, 
scent, and colour materials, together with a 
general description of the special structural adap- 
tations for floral diet to be traced in the different 
groups of insects. 
The body of the work (pp. 90-385) is devoted 
to examples of floral mechanism of different types 
of flowers; but since the number of flowers avail- 
ing 
able is so enormous, and the adaptations so in- 
finitely varied, some method of selection is 
desirable (a review of all species and genera in 
the province of the 
Kirchner has solved 
systematic order being 
volumes of Knuth). Prof. 
NO. 2218, VOL. 89] 
the problem in an ingenious manner by adopting 
the flower-classes of Miiller, and selecting about 
ten examples of each class for more detailed ex- 
amination and illustration: the value of the work 
consists largely in the breadth of view exercised 
in the selection of suitable forms. Thus the 
flower-classes of Miller, indicated by the irrita- 
ting symbols PO, A, AB, B, B’, D, H, F (these 
symbols being replaced by others in the English 
translation of Knuth), are treated in separate 
chapters, though Class H (Hymenoptera-flowers) 
extends to eight sections, of almost equal value, 
each of which is again illustrated by a series of 
forms. The types, it is interesting to note, are 
selected for their intrinsic value, quite apart from 
their connection with a comparatively poor in- 
digenous flora. Flowers in general garden or 
greenhouse cultivation, as well as types from 
south Europe, or even further atield, are called 
upon to assist in creating an impression of the 
scope of the subject. 
It is interesting to note that the Crocus flowers, 
which with us only open for a few hours in warm 
sunshine, are correlated, particularly the white 
ones, with the visits of night-flying moths in the 
Alps, while autogamy in this case is effected in 
older flowers by a continued growth of the 
perianth-tube, which enables the stamens to make 
contact with the stigmas. Such. inconsistencies, 
however, are useful as stimulating observation on 
the part of the reader. 
The end chapters are devoted to the discussion 
of examples of the application of the statistical 
methods developed by Loew, McLeod, and 
Robertson, and the manner in which they may be 
utilised to throw light on the special ecological 
characteristics of the flora of a given climatic 
region; to the problems of the fundamental cause 
of the existing mutual and complementary rela- 
tions between plants and their pollinating guests; 
and, finally, to the present state of our knowledge 
with regard to the origin of the Angiosperm 
flower and its visitors mainly from the point of 
view of geological evidence. 
COSMOGONIES, OLD AND NEW. 
Lecons sur les Hypothéses Cosmogoniques 
Professées &@ la Sorbonne. By H. Poincaré. 
Rédigées par H. Vergne. Pp. xv+294. 
(Paris: A. Hermann et Fils, 1911.) Price 12 
francs. 
HE most valuable part of the book under 
Bb review is undoubtedly the preface. Here 
M. Poincaré expresses something of his own 
The imperious claim upon our minds of 
cosmogonical speculation is first justified, then a 
views. 
