Collins: Nature of Mendeltan Units 



437 



factors, as we now understand them, 

 are ultimate and indivisible or that they 

 originated by single steps. 



COMPLEXITY OF CHARACTERS. 



Even the simplest visible character is 

 of such a complex nature and requires 

 the interaction of so many elements 

 that the simultaneous origin of the 

 indi\ddual elements, all properly cor- 

 related, is difficult of conception. The 

 possibilities of subdivision are not 

 limited, even to visible factors. Analy- 

 tical breeding has shown that many of 

 the elements of a visible character, i^ 

 they exist alone, are entirely invisible. 

 Factors for pattern and intensifying 

 factors may be combined, segregated, 

 and recombined, and yet give no visible 

 evidence of their existence until united 

 with the necessary factor for pigment. 

 In many instances this subdivision into 

 factors continues to the limit of our 

 ability to analyze, and beyond this point 

 we have no way of knowing to what 

 degree of minuteness subdivision may 

 extend. One result of this class of in- 

 vestigations is to bring to our minds a 

 reahzation of the multitude of diversi- 

 ties that actually exist, but concerning 

 which our senses give us no direct 

 intimation. 



The problem of evolution has been 

 attacked from one side by biometricians 

 and others adhering to the idea of 

 gradual change, and from the opposite 

 side by the followers of De Vries and 

 Mendel. The field between them has 

 been narrowed until the two forces are 

 now within sight of each other. There 

 are still many points of divergence, but 

 all that remains of the question as to 

 whether or not evolution proceeds by 

 abrupt steps is a different terminology 

 and a different point of view. If the 

 individual steps by which evolution 

 proceeds are too small to be distin- 

 guished, the question whether evolution 

 should properly be termed continuous 

 or discontinuous is chiefly of academic 

 interest. 



The futility of the distinction be- 

 tween blended and discontinuous in- 

 heritance has been appreciated by 

 Reid:^ "Adherents of the Mendelian 



doctrine sometimes maintain that 

 blended inheritance is nothing other 

 than a fine mosaic. Doubtless they are 

 right; but I imagine their opponents 

 mean nothing more by blended in- 

 heritance than that it is a very fine 

 mosaic — a mosaic as fine, perhaps, as 

 the mixture of colours in a painter's 

 brush. All blending, for example a 

 mixture of alcohol and water, is a mosaic, 

 even if one of chemical molecules. " 



With the idea of illustrating how this 

 common viewpoint may be applied to 

 characters that behave as simple Men- 

 delian units, the evolution of a hypothet- 

 ical character may be traced. 



In the palms, the pinnate type of 

 leaf is held to have developed from forms 

 with palmately divided leaves. In the 

 typical palmate leaf, such as Chamae- 

 rops, the veins all arise from the tip of 

 the petiole. Let us assume that in a 

 geographically segregated group, the 

 tip of the petiole or point of origin of 

 the veins becomes gradually elongated. 

 This elongation continues urtll the 

 pinnae are finally separated and the 

 pinnate type of leaf is attained. The 

 magnitude of the individual steps by 

 which this development procedeed, we 

 assimie to have been as small as the 

 differences now observed between in- 

 dividuals of existing species. If now a 

 palm of the pronounced pinnate type, 

 such as Cocos or Phoenix, could be 

 crossed with a Chamaerops, the type of 

 leaf would doubtless Mendelize, perhaps 

 without intermediates. It would then 

 be assumed that the difference in leaf 

 form was represented by a unit character 

 and that this difference represented a 

 single step in evolutionary progress. If, 

 however, some species with a third type 

 of leaf should come to light, say such a 

 plant as Sabal, that has a leaf inter- 

 mediate between the palmate and pin- 

 nate types, and this new form could be 

 crossed with the Chamaerops or Phoenix, 

 it would then be found that the leaf 

 character separating Chamaerops from 

 Phoenix consisted of two factors, each 

 of which might still be found to Men- 

 delize, thus halving the apparent evolu- 

 tionary step. Further subdivision of 

 the character would depend on whether 



2 Reid, G. A., "The Laws of Heredity," P. 155, footnote. 



