ON THE MEALY LINNET. 457 



those of our native bird, «ven allowing for a moderate degree of variance more 

 than is observable — the birds themselves appear to be quite undistinguishable. 



We have now seen each character, in its turn, stand forth as the sole exception 

 to complete resemblance, though reduced to the very minimum of non-conformity ; 

 and this in races inhabiting the same districts, as well as in those which people 

 different parts of the globe, — thus negativing the assumption that external 

 influence had necessarily produced the diversity, which theory obviously requires 

 for its support the existence of intermediate specimens in all their progressive 

 stages, and is consequently inconsistent with the observed permanency of the 

 various races in question. At the same time, as I have remarked on a former 

 occasion, although these closely-allied races, when inhabiting the same locality, 

 apparently manifest no disposition to intermix and blend, as is uniformly the 

 case with even the most distant of true varieties, such as are caused by domes- 

 tication — and it may be that many such races constantly keep apart, even though 

 so similar that we possess no means of distinguishing them — yet that, in the 

 case of an occasional union (brought about by peculiar circumstances), it is highly 

 probable that the tendency to barrenness in the mule progeny would decrease 

 in the ratio of the amount of physiological affinity subsisting between the parents, 

 until in the most approximate races (though not descended from a common origin, 

 and therefore specifically distinct) the hybrid offspring would be even mutually 

 as prolific as individuals of unstained descent. In connexion with this specula- 

 tion, the highly curious physiological fact may be borne in mind — which possibly 

 may have for its object the continued separate existence of each primitive 

 race — that very closely-allied species, and (as an apparent consequence of the 

 same law) similar varieties also, less frequently produce, on intermixture, offspring 

 of blended character, than progeny entirely resembling one or the other parent. 



However, having launched into conjecture, it may, on the other hand, be sup- 

 posed, as perfectly consistent with this presumed law, adjudged to regulate the 

 productiveness of hybrids, that the aversion to interbreed would lessen with each 

 additional degree of affinity ; till in the most approximate races of all, though 

 still not descended from the same original parentage, all mutual unwillingness to 

 reproduce would be entirely extinguished, — so that, in a few generations, nothing 

 would prevent their becoming effectually blended, save that inexplicable tendency 

 (so remarkably evinced by even emigrants) to return to pair and propagate in the 

 native locality, which must necessarily occasion relatives to transmit their race in 

 far greater proportion than individuals of diverse lineage, — a circumstance which 

 it is truly surprising does not induce a much greater splitting into varieties than 

 is actually observed ; and this, if climate or locality really exercised the amount 

 of influence which many naturalists imagine, would confer distinctive characters 

 on the individuals of almost every province. 



