494 GENETICS IN RELATION TO AGRICULTURE 
of performance. Selection has indeed been a powerful factor in the 
improvement of modern breeds of livestock, but it is not necessary to 
assume that selective improvement is due to anything other than the 
isolation and multiplication of animals possessing the best combinations 
of germinal elements. Mutations or germinal changes sometimes occur. 
They are not necessarily favorable, but when useful they are likely to 
be preserved. They add to the store of heritage in our breeds of livestock. 
Favorable conditions of this kind may be maintained, and to the non- 
critical mind produce results simulating the inheritance of acquired 
characters. 
The Conclusion.—There appears to be no escape from an attitude 
of extreme scepticism with respect even to the transmission of functional 
modifications. There is no conclusive experimental demonstration in 
the true sense of the word of the inheritance of any acquired character, 
and there is abundant experimental evidence, although it cannot of 
course be conclusive, for the contrary belief. Accordingly it should be 
evident to the practical breeder that anything which is so difficult for 
scientists to demonstrate must have little possibility of practical value. 
There is enough experimental evidence to demonstrate that it cannot 
possibly be the rule for the great majority of characters, and that it 
cannot produce significant effects in short times. The individual breeder 
works with relatively few generations, and he should shape his operations 
in accordance with that fact. 
But although the inheritance of acquired characters may be denied, 
particularly from a practical standpoint, that denial does not carry with 
it any under-estimation of the importance of modifiability in animal 
breeding operations. Thomson has given this point happy expression 
in the statement that, “ Although what is ‘acquired’ may not be inherited, 
what is not inherited may be acquired.” And also just as some of 
the data of practical breeding operations seem to indicate an inheritance 
of acquired characters, so some methods of breeding the success of which 
is apparently based upon the inheritance of acquired characters really 
depend for their success upon harmony with other laws of heredity. 
A change in theoretical interpretation need not necessarily change 
breeding methods. , 
Digitized by Microsoft® 
