84 INTRODUCTION. 



is often possible to infer from an isolated organ or structure the es- 

 sential characters of the remainder of the organism. Thus, if we were 

 acquainted with no other part of some animal than its skull alone, 

 and if we found that that skull possessed two occipital condyles, and 

 that each half of the lower jaw was composed of a single piece, we 

 should be justified in concluding that the animal to which the skull 

 belonged possessed mammary glands. We should also be justified 

 in inferring many other facts about it, as, for example, that it pos- 

 sessed (or might have possessed) a hairy covering to the body, that 

 its blood was hot, and that it possessed non-nucleated red blood- 

 corpuscles. Similarly, if we met with a mammalian lower jaw, the 

 angle of which was bent inwards, or "inflected," we should have 

 a strong presumption that the animal to which it belonged possessed 

 " marsupial bones " or cartilages on the brim of the pelvis, and that 

 the young were born in a very imperfect state of development. It 

 follows from what has been already said, that the law of the correla- 

 tion of organs plays a most important part in palaeontological inves- 

 tigation, enabling the observer to more or less completely "recon- 

 struct " an extinct organism by means of its fragmentary remains. It 

 is to be remembered, however, that the law is a purely empirical one, 

 and expresses nothing more than the result of experience ; so that 

 structures which we now know only as occurring in association may 

 ultimately be found separate, and conjoined with structures of a 

 different character. 1 Moreover, it is to be borne in mind that in any 

 two correlated structures it is not that each is correlated with the 

 other, but that one of the two is correlated with the other. That is 

 to say, of any two correlated organs, A and B, it may be true that A 

 is never found without B, but it does not follow that B may not 

 occur without A. Thus, the presence of a stomach adapted for 

 " rumination " is invariably associated (in living types) with an im- 

 perfect development of the incisors of the upper jaw, the central 

 upper incisors being always wanting ; but it is not the case that an 

 incomplete condition of the upper incisors, or the absence of the 

 central ones, is necessarily correlated with the habit of chewing the 

 cud. The proper way of putting the case is to assert that certain 

 structures (A) are never found apart from other structures (B), though 

 the latter may be present without the former. When, therefore, we 

 find a lower jaw having its angle " inflected," we may, with our 

 present knowledge, assert that the animal to which that jaw belonged 

 probably possessed " marsupial bones " or " marsupial cartilages " 

 upon the brim of the pelvis ; although the presence of a certain 

 amount of inflection in the jaws of some Insectivora would preclude 



1 A remarkable instance where this correlation is at fault, and has led to the 

 reference of the bones of one animal to two distinct orders, will be noticed 

 among the Mammalia under the head of Chalicotherium. 



