of \itnl iiilicrilancc. 'I'lic chief point is, that they arc t\|H' idials. transmit ii-J 

 nnd preserved solely through the form relation. 



The olijection may be urged that any condition, whicli wiiii rigid exactness 

 preserves the continuity of type, precludes the possibility of variation, upon 

 which tendency all evolutionary development denpends. Such criticism, how- 

 ever. g\-erlooks the fact that the form-tendency \yliich expresses itself in the 

 phenomena of variation is ever present in the species as a characteristic of the 

 type, and may remain latent and unmanifestcd indclinitely, if favorable condi- 

 tions do not ari.se to induce its expression. The form-tendency to vary falls 

 under the same mathematical law as the preservation of type. No mathematical 

 principle could be more rigorous. l'"very parabola, or hyperbola, is an illustration 

 of the necessary and inflexible order of hereditary deflection. 



'I'hc specialized forms directly concerned in heredity may be regarded as 

 combinations of elemental electronic groupings. It may well be, therefore, that 

 these higher, synthetic forms are subject to laws similar to those which govern 

 electronic charges, varying in stability, breaking down suddenly, and re-com- 

 bining in a new group formation. If the conditions and processes determining 

 the form-character of the factors in heredity are but extensions and vital 

 adaptations of laws governing elemental electronic phenomena, the origin of 

 species by mutation, or saltation, is just what we should expect. The occasional 

 appearance of a "sport" in the direct line of inheritance would indicate that the 

 number of group-units constituting the character-form had been, perhaps slightly, 

 increased or descreased. when the formation was already at a high state of 

 instability. The sudden breaking down of the ancestral formation and the 

 re-arrangement of the constituent parts in a new form, indicate the same pro- 

 cesses that are observed in electronic phenomena. 



If we closely follow the electronic analogy, the suddenly created new 

 character, or new species, would be extremely unstable relati\-e to one of two 

 possible conditions. If the critical number of units necessary for the new 

 grouping was obtained by the addition of a single unit, the new character 

 would be extremely unstable relative to any loss thereform; for, in that case, 

 the loss of a single unit would result in the sudden extinction of the new 

 character, and the reversion to the ancestral form. The entrance of additional 

 units would continue to increase the stability of the new form up to a certain 

 point, beyond which, if the process of addition continued, the stability would 

 decline to a new breaking point. 



On the other hand, if the new character resulted from the loss of a single 

 unit from the ancestral form, the new creation would be subject to the exact 

 reverse of these conditions. The loss, then, of one or more units would render 

 the form more stable up to a certain point, after which, were the process con- 

 tinued, the stability would decline to a new point of dissolution. This form, 

 however, could not receive a single additional unit without breaking down and 

 reverting to the ancestral character. 



When a character created by unit addition breaks down through the con- 

 tinued increase of its constituent units; or when a character resulting from unit 

 loss suddenly disappefirs through the reverse process; the new character which 

 then arises is something essentially different from the parent "sport," and from 

 their common ancestor. 



If, in cultural propagation, it is desired to maintain fixedly a certain sport 

 character, it is necessary that the form be kept as nearly as possible at its 

 highest point of stability. What this point, in any given case, may be, and 

 what may be the best methods by which to reach it, are in the very nature of 

 the case difficult to determine. Careful experimentation, however, might deter- 

 mine whether its stabilit}^ Is on the side of unit addition or unit loss. In 

 either case, cross-fertilization In a certain direction. If selection Is wisely made,, 

 may secure Increased stability. In the absence of anything better, direct-line 

 breeding may keep the character fixed; but In such case It should always be 

 remembered that the character Is stable from one side only, and always extremely 

 unstable In the other direction. Possibly, through selective propagation only, 

 a unit may occasionally be added to or taken from the character form to 

 Increase its stability; but in no case can such a process continue Indefinitely, 



