DEPARTMENT OF EXPERIMENTAL EVOLUTION. 8/ 



same result on all developing somas, but, on the contrary, the result varies 

 with the responsiveness of the soma and this is a function of the germ-plasm. 

 A large series of investigations into the laws of fluctuations and the relations 

 of fluctuations to varying nutrition are being carried out by Dr. J. Arthur 

 Harris at this Station. 



Another error in the soma as an index of the characters of the germ-plasm 

 is possible whenever the germ-plasm contains the potentiality for a low 

 degree, or the absence, of a quality as well as for a high degree. A par- 

 ticular soma will frequently show only the higher degree — the lower de- 

 gree is veiled by the higher. This error may be corrected by the study of 

 a large number of somas from the given germ-plasm, since some of these 

 will lack the higher quality and reveal the lower. Thus, pigmented flowers 

 or hair may veil a potential albinism in the germ-plasm, but a certain pro- 

 portion of the offspring will, under proper breeding, be albinos. Again, it . 

 frequently happens that two or more factors are required to produce a visible 

 result. If one factor alone is present in the germ-plasm it will not be revealed 

 in the soma. The germ-plasm first reveals this factor by its behavior when 

 the missing factor is added by crossing. Thus, the germ-plasm of the yellow 

 canary has a pattern but no pigment to reveal it in the soma. When to this 

 germ-plasm is added the pigment-forming quality of the "green" canary, or 

 of some other species of finch, the pattern is revealed in the soma of the hy- 

 brid. From a consideration of the foregoing facts it is plain that a single 

 soma does not afford a complete analysis of its germ-plasm ; but by using spe- 

 cial methods many somas may, together, be made to reveal the complete con- 

 tent of the germ-plasm. 



Using, with proper care, the soma as the analysis of its germ-plasm, we 

 seek to study the origin of changes in germ-plasm. First, we have paid atten- 

 tion to cases where, it is alleged, the germ-plasm is undergoing a sort of nat- 

 ural, one might say spontaneous, change. The most famous case of this is 

 that of the evening-primrose, to which de Vries called attention. We are 

 breeding evening-primroses extensively, and some of the results of our work 

 have been published in conjunction with studies made by Dr. MacDougal.* 

 The capacity of this wonderful plant for inciting investigations upon itself 

 seems unlimited. We have only begun an era of investigation into its germ- 

 plasm ; and in this plant the structure of the germ-plasm of different species 

 exhibits peculiarities that can readily be seen with the microscope. Miss 

 Anne M. Lutz has devoted much time to the study of the stainable bodies 

 (chromosomes) of the various species and has found the number to vary 

 greatly, as, e. g., 14, 15, 28, 29, 30. When a species with 15 chromosomes is 

 crossed by one having 30 the hybrid offspring have 22, being half-way be- 

 tween the parental- numbers, but a few show the ancestral group numbers 



* Carnegie Institution of Washington, Publications Nos. 24 (1906) and 81 (1907). 



