76 SECTIONAL ADDRESSES. 



hardly be detected on the surface of the skull ; {b) there is some pre- 

 determining law or similarity of potential which governs their first 

 existence, because (c) the rudiments arise independently on the same 

 part of the skull in different phyla [i.e. hneages] at different periods of 

 geologic time; (d) the horn rudiments evolve continuously, and they 

 gradually change in form (i.e. allometrons) ; (e) they finally become the 

 domanating characters of the skull, showing marked variations of the 

 form in the two sexes ; (/) they first appear in late or adult stages of 

 ontogeny, but are pushed forward gradually into earlier and earUer 

 ontogenetic stages until they appear to arise prenatally. ' 



Osborn says that rectigradations are a result of the principle of 

 detennination, but this does not seem necessary. In the first place, 

 the precise distinction between an allometron and a rectigradation fades 

 away on closer scrutiny. When the rudiment of a cusp or a horn 

 changes its form, the change is an allometron; the first swelling is a 

 rectigradation. But both of these are changes in the form of a pre- 

 existing structure ; there is no fundamental difference between a bone 

 with an equable curve and one with a slight irregularity of surface. 

 Why may not the original modification be due to the same cause as the 

 succeeding ones? The development of a horn in mammalia is prob- 

 ably a response to some rubbing or butting action which produces 

 changes first in the hair and epidermis. One requires stronger evidence 

 than has yet been adduced to suppose that in this case form precedes 

 function. As Jaekel has insisted, skeletal formation follows the changes 

 in the softer tissues as they respond to strains and stresses. In the 

 evolution of the Echinoid skeleton, any new structures that appear, 

 such as auricles for the attachment of jaw-muscles or notches for the 

 reception of external gills, have at their inception all the character of 

 rectigradations, but it can scarcely be doubted that they followed the 

 growth of their cori'elated soft parts, and that these latter were already 

 subject to natural selection. But we may go further: in vertebrates 

 as in echinoderms the bony substance is interpenetrated with living 

 matter, which renders it directly responsive to every mechanical force, 

 and modifies it as i-equired by deposition or resorption, so that the 

 skeleton tends continually to a correlation of all its parts and an adapta- 

 tion to outer needs. 



The fact that similar structures are developed in the same positions 

 in different stocks at different periods of time is paralleled in probably all 

 classes of animals ; Ammonites, Brachiopods, Polyzoa, Crinoids, Sea- 

 urchins present familiar instances. But do we want to make any 

 mystery of it? The words 'predisposition,' ' predetennining law.' 

 'similarity of potential,' 'inhibited potentiality,' and 'periodicity,' all 

 tend to obscure the simple statement that like causes acting on like 

 material produce like effects. When other causes operate, the result is 

 different. Certainly such facts afford no evidence of predetei'mination, 

 in the sense that the development must take place willy-nilly. Quite 

 the contrary : they suggest that it takes place only under the influence 

 of the necessary causes. Nor do they warrant such false analogies as 

 ' Environment presses the button : the animal does the rest. ' 



The resemblance of the cuttle-fish eye to that of a vertebrate has 



