CONCLUSIONS 811 



(e) The expression of these characters in these Devonian samples of 

 the race is clearly intermediate between their expression in typical sam- 

 ples of the race taken from the Silurian and from the Carboniferous 

 respectively. 



(/■) It appears reasonable to generalize these facts to form the state- 

 ment that phylogenic modification follows the law of ontogenic growth. 



This appears to me, after making these studies, to be a more correct 

 form of statement tlian to say, as we have been accustomed to do, that 

 the ontogeny is a recapitulation of the ph^dogeny. From the facts it 

 seems more probable that the direction of evolution is determined by the 

 laws of individual growth than the converse, so far as there is any rela- 

 tion between the two. 



(g) The prominent fact thus brought out, namely, the persistency of 

 fluctuation of characters for such a long line of descent, suggests a new 

 conception of the nature of organic variation. 



The behavior of fluctuating variations indicated by this study of Ehipi- 

 domella. indicates that the fluctuating is probably an effect of the fluctu- 

 ating' environment reacting on characters which are transmitted without 

 fluctuation. 



To use an illustration : Suppose a number of buildings were constructed 

 according to a definite plan, in Avliich the exact number and order of 

 arrangement of the bricks of the walls was determined. The houses thus 

 constructed would be alike if the bricks were of rigid material; but if 

 the bricks were made of rubber or elastic material, the lines of the house 

 would constantly vary with the load put on them. The conception sug- 

 gested by these facts is that the plan of construction for each organism 

 is transmitted by heredity with the same degree of definiteness for fluctu- 

 ating as for unfluctuating characters, but that the reason why some 

 characters are fluctuating is because the materials of construction are 

 susceptible to the influence of environment during growth, and hence 

 reflect directly the influence of the fluctuating stresses of environment. 



(h) This elasticity of the materials of construction enables the organ- 

 ism to adjust itself to different environment in its growth, but the adjust- 

 ment does not in any degree affect the plan of construction or whatever 

 it may be which is concerned in the transmission of characters in heredity. 



(i) This variability is akin to the motion of movable parts of an organ- 

 ism, and concerns hard parts only where they come into direct contact 

 with environment or with active (muscularly) soft parts of the structure. 



(/*) These fluctuating characters are as important to the organic econ- 

 omy of the organism as are the rigid, hard parts, and are as essential in 

 the constitution of the species ; as it might be said that the joint of the 



