THE PROBLEM OF ORGANIC DEVELOPMENT. 181 



faith in the old transmission theory. Richard Semon's recent work, "The Memory 

 as the Conservative Principle in Evolution," appeals to admiration as a splendid 

 contribution in behalf of the same defunct theory. 



The principal theories now in the field are three: selection (Darwin and Wallace) ; 

 mutation (de Vries); orthogenesis (Niigeli and others). 



The only tests for theories are, of course, facts; and by facts I mean not such 

 facts as we may compile by the hundreds, when we are merely counting variations 

 the relations and origins of which we only know by report, or by hurry-scurry 

 observation, but facts that can be examined in their natural connections, facts that 

 have an approachable history and are open for consecutive study and experimental 

 handling. The worst of all faults in research is to snatch at facts and ignore their 

 genetic relations. 



In order to show how selection, mutation, and orthogenesis differ in attitude 

 toward facts I now bring before you a single brief study in the transmutation of a 

 color-pattern. It is a study in specific characters that admit of being traced step 

 by step in the living bird and that are amenable not only to comparative and genetic 

 study but also to simple experimental test. It is a study which will also illustrate 

 the biogenetic law — continuity in the processes of specific development, actual flow- 

 ing transitions between characters which, at first sight, appear to be separated by 

 mutational discontinuities. The importance of such transitions is at once apparent 

 when we recall that the mutation theory is based on the assumption of discontinuous 

 unit-characters, between which there can be no real transitions. 



Possibly some who are not familiar with color-marks as specific characters may 

 feel, as we have elsewhere remarked, that they have not the same genuine tangible 

 specificity as is exhibited in bones, tissues, and organs. Colors may fade and pass 

 through shadowy phases that to the casual observer give them an air of insubstan- 

 tiality. Then, too, there is such profusion and intermingling of colors, especially 

 in domesticated animals and plants, that they are apt to impress us as emblematic 

 of inconstancy. 



When, however, we turn to wild species, we get a very different impression. Here 

 color-characters are as constant as other characters, and as they are displayed at 

 the surface, they have the great advantage of accessibility for study and experiment. 



One other point may be briefly touched here. Prevailing methods of collecting 

 data, especially in the field of evolution, have tended to overmagnify the importance 

 of mass and large numbers and to underestimate the value of the all-round study 

 of the individual case. The pressing need of the hour, as I see it, is the persistent 

 exhaustive study of the single favorable cases. Crucial evidence is always individual 

 in the last analysis. Many half-truths can not make a whole truth, and the omni- 

 vorous rambling collector should learn from Darwin, Mendel, and de Vries how 

 much more effective attention becomes when it dwells intensively on some advan- 

 tageous focal point. 



Let this be my apology, if one be needed, for inviting attention to a single color- 

 pattern, as one among many available cases in which to point out a test for theories. 



Our three theories again briefly characterized are as follows: 

 (1) Mutation stands for gaps and no bridges — jumping the very problem to be 

 solved. Its discontinuities are unpredictable breaks in the lines of derivation, pre- 

 cluding both foresight and hindsight. 



