216 CORRESPONDENCE (September 1898 
and because conditions are usually more powerful, and necessarily operate for shorter 
periods of time, when under man’s direction than in nature, so variations should be 
correspondingly less definite in the domesticated forms of life. 
This position also destroys the force of the two chief arguments used against Natural 
Selection by demonstrating that it is itse/f able to induce those favourable variations, which 
it subsequently selects from. Firstly, it is urged that Natural Selection being unable to 
produce definite variations must be dependant on some other factor until the 
variations are sufficiently far advanced to be of selective value, and it is therefore in- 
competent to solve the most important question in the formation of species ; secondly, 
the number of coincidences, which are necessary to perpetuate any given favourable varia- 
tion on the assumption of indefinite variability, are so great that Natural Selection must 
be regarded, at best, only as a subordinate factor. By showing that, if natural selec- 
tion acts at all, it must tend to produce definite variations, these two objections are 
largely overcome, and the facts adduced by the Neo-Lamarckian school easily accounted 
for. Hence with two competing theories to explain species formation, it becomes neces- 
sary to makea further appeal to facts to determine the value of each. 
I feel so convinced of the importance of this aspect of the subject that I should be 
sorry if any want of clearness on my part at all obscured the point at issue. 
J. LioNEL TAYLER. 
THE GrorTo, HAMPTON-ON-THAMES. 
NOTICE 
To Conrrisutors.—All Communications to be addressed to the Epiror of NATURAL 
Screncr, at 29 and 30 Bedford Street, London, W.C. Correspondence and Notes 
intended for any particular month should be sent in not later than the 10th of the 
preceding month. 
To THE TRADE.—NATURAL SCIENCE is published on the 25th of each month ; all 
advertisements should be in the Publishers’ hands not later than the 20th. 
To our SUBSCRIBERS AND OTHERS.—There are now published TwELve VoLUMEs 
oF Natura Science. Nos. 1, 8, 11, 12, 13, 20, 23, 24 being OUT OF PRINT, can only 
be supplied in the set of first Four Volumes. All other Nos. can still be supplied at 
ONE SHILLING each. 
Price of Set of Vols. I., II., III., IV. - : 5 ose 10) x0 
Ne Wie Vid va: 5 : rs ke 
I.—VIII. : ; ; : Zee 10.70 
99 39 ”” 
2? 3) 99 
