622 REPORT— 1889. 



atrophy or hypertrophy at certain points. (3) The variations thus arising are 

 accumulated by the selection of the individuals in which they are most marked and 

 by the extinction of inadaptive natural groups. Selection, in so far as it affects 

 these variations, is not of single characters but of the ensemble of characters. 



The evidence is of a direct and indirect character. The direct evidence is that 

 by actual observation in complete palseontological series, the origin of adaptive 

 structures is found to conform strictly to the lines of use and disuse. The 

 indirect proof is that the natural selection of chance variations is unsupported by 

 observation and is inadequate to explain the variation phenomena of the second 

 class. 



4. I will first briefly consider the former. The distinctive feature of palceonto- 

 logical evidence is that it covers the entire pedigree of variations, the rise of useful 

 structures not only from their minute, apparently useless, condition, but from the 

 period before they appear. The teeth of the mammalia render us the most direct 

 service, as compared with the feet, since they furnish not only the most interesting 

 correlations and readjustments, but the successive addition of new elements. With 

 a few exceptions which need not be noted here, all the mammalia started with teeth 

 of the simple conical type — like the simple cusps of reptiles. Practically every 

 stage between this sjngle cusp and the elaborate multicusped recent molars is now 

 known. Every one of the six main cusps of the molar of Hyracotherium, for 

 example, a type of an important central stage in the ungulate dentition, is first 

 indicated at the first point of contact or extreme wear between the upper and lower 

 molars ; this point of wear is replaced by a minute tubercle, which grows into a 

 prominent cusp. These are the laws of cusp-development, as observed in every 

 known phylum of mammalia : — 



I. The primary cusps first appear as cuspules, or minute cones, at the first 

 points of contact between the upper and lower molars in the vertical motions of 

 the jaws. 



il. The modelling of cusps into new forms, and the acquisition of secondary 

 position, is a concomitant of interference in the horizontal motions of the jaws. 



5. The evidence, of which this is only a single illustration, has accumulated 

 very slowly. The line of reasoning from this particular series of observations is 

 as follows : — 1. The new main variations, in the teeth and skeleton of every com- 

 plete series, are observed to follow certain definite purposive lines. 2. By careful 

 analysis of the reactions to environment which would occur in the individuals by 

 the laws of growth — we observe that the race variations strictly conform to the 

 line of these reactions. 3. We further observe that no variations of this class 

 occur without the antecedent operation of these reactions ; the working hypothesis 

 thus stands the test of prediction. 4. We accept this invariable sequence of race 

 adaptation upon individual adaptation as proof of a causal relationship. 



6. I admit that this proof may be invalidated in several ways : — 1. By show- 

 ing in more extended research that these observations of sequence are inaccurate 

 or offset by others in which there is no such sequence. 2. By showing that the 

 Lamarckian principle, while explaining some of the variations of this class, ia 

 directly contradictory to others. 3. By showing that all these phenomena may be 

 explained equally well or better by natural selection. 4. By proving, indepen- 

 dently, that the transmission of acquired characters never occurs. 



I will now consider each of these cases : — 



1. As regards these observations. — They may be examined in detail in the 

 studies of Cope, Wortman, or Ryder, and in a paper I presented to this Associa- 

 tion last year. As the question of transmission has been generally assumed in the 

 foregoing studies, I think it is now important to review the whole field, searching 

 for facts which look against the Lamarckian principle ; for as we have been hitherto 

 studying with a bias in favour of it, some such adverse points may have been over-' 

 looked. At present, however, I can recall only a single adverse observation, that 

 is, in the development of one of the upper cusps, the lower cusp which opposes it, 

 and which is therefore supposed to stimulate this development, is foimd to recede. 

 I have no doubt others will be found presenting similar difficulties. 



2. As regards the Lamarckian principle. — Several objections to the special 



