568 FIXITY OF TYPE IN THE BREED OF SHEEP. 
France, that an opposite state of things had obtained in all 
these trials ; since purity and antiquity of blood exist much 
more strongly in the French breeds than in the English, 
which have been much more recently formed. The imperfect 
result then of all these attempts is completely accounted for 
by our reversal of a great law of nature ; and it seemed to me 
necessary to restore this law and give the advantage of it to 
the English ram. Such was the preliminary condition of 
success. 
It appeared then that in order to untie the Gordian knot 
whose threads I have traced, inasmuch as one could not 
increase the purity and antiquity of the blood of the rams 
(I purposely repeat the first principles of the problem to be 
solved), one must diminish the resisting power, namely, the 
purity and antiquity of the ewes. With a view to this new 
experiment, one must procure English rams of the purest 
and most ancient race, and unite with them French ewes of 
modern breeds, or rather of mixed blood forming no distinct 
breed at all. It is easier than might be supposed to 
combine these conditions. On the one hand, I selected 
some of the finest rams of the New-Kent breed, regenerated 
by Goord. On the other hand, we find in France many 
border countries lying between distinct breeds, in which 
districts it is easy to find flocks participating in the tw 7 o 
neighbouring races. Thus, on the borders of Berry and La 
Sologne one meets with flocks originally sprung from a 
mixture of the tw 7 o distinct races that are established in those 
tw 7 o provinces. Among these then 1 chose such animals as 
seemed least defective, approaching, in fact, the nearest to, 
or rather departing the least from, the form w 7 hich I wished 
ultimately to produce. These I united with animals of 
another mixed breed, picking out the best I could find on the 
borders of La Beauce and Touraine, which blended the 
Tourangelle and native Merino blood of those other two 
districts. From this mixture v T as obtained an offspring com- 
bining the four races of Berry, Sologne, Touraine, and 
Merino, without decided character, without fixity, with little 
intrinsic merit certainly, but possessing the advantage of 
being used to our climate and management, and bringing to 
bear on the new breed to be formed, an influence almost 
annihilated by the multiplicity of its component elements. 
Now, what happens when one puts such mixed-blood ew r es 
to a pure New-Kent ram? One obtains a lamb containing 
fifty hundredths of the purest and most ancient English blood, 
with twelve and a half hundredths of four different French 
races, which are individually lost in the preponderance of 
