974 THE AMERICAN NATURALIST, [VoL. XXXIII. 
discussed under the fertilization of the egg, but no practical work 
is suggested. Some of the objects suggested for study are not of 
sufficiently wide or abundant occurrence to make them generally 
accessible, e.g., the eggs of Thysanozoon, Sida, Myzostoma, Siredon, 
and the repetition of Boveri’s experiment on the fertilization of enucle- 
ated egg fragments. In spite of these limitations, the work will be of 
great value to all who are giving courses on practical cytology. 
Davenport’s Statistical Methods.1— This work is intended to 
meet “the call for a simple presentation of the newer statistical 
methods in their application to biology,” and seems an admirable 
handbook for the purpose. It consists of definitions and explana- 
tions of methods, including “the seriation and plotting of data and 
the frequency of the polygon,” etc., with a selected bibliography of 
the subject, the whole occupying less than fifty pages, followed by 
about one hundred pages of formulas and logarithmic tables. Chap- 
ter V (pp. 38, 39) gives “Some Applications of Statistical Biological 
Study.” While the “newer statistical methods” are admirably 
suited to the investigation of certain special problems, which may be 
of the highest interest and importance, they seem too minute and 
detailed, and to require the expenditure of too much time and labor, 
to be of very broad application, such as the author apparently 
contemplates. Thus, it is said: “ The origin of species through geo- 
graphical segregation can be studied by the determination of A/ace- 
modes ; that is, the modal condition of specific characters of one and 
the same species in various localities. The progress of specific dif- 
ferentiation will be measured by the change in place-modes from 
decade to decade, or by the formation of a binomial curve in the 
place of a modal one; by the gradual separation of the two modes 
of a binomial curve.” Theoretically this is possible, but taking into 
account what it implies, even for a single species, does it not border 
on the chimerical, or at least on the impracticable? First is neces- 
sarily involved a geographical area of considerable extent — at least 
hundreds of miles square, under ordinary conditions of topography 
— within which many observation stations must be chosen, and 
at which work must be continued “from decade to decade,” and 
detailed measurements made of every measurable feature of many 
1 Statistical Methods with special reference to Biological Variation. By 
C. B. Davenport, Ph.D., Instructor in Zodlogy at Harvard University. First 
edition. First thousand. New York, John Wiley & Sons. London, Chapman 
& Hall, Limited, 1899. — 12mo, 148 pp. 
