Studies of Teratological Phenomena. Ill 



important question theoretically, on account of its bearing on the theory 

 of gaoietic contamination as opposed to that of gametic purity. My 

 casual observations on the 402 race are not in a sufficiently definite 

 form as yet to convince others than students of this same material that 

 they are accurate. This being true, two interpretations can be placed 

 upon my data. One may say, until proof to the contrary in a tabulated 

 form is presented, that segregation of fasciation- determining material 

 was not complete in the formation of the F2 seed, and consequently 

 slight abnormalities in aa flowers appear, the latter being evidence in 

 favor of such a supposition. However, one may also say that these 

 abnormal flowers occur just as frequentlj^ in the 402 race, that they 

 are just as abnormal, and that their presence is the result of accidents^) 

 in ontogeny, in which case, internal character-materials have undergone 

 absolute segregation and the F2 aa plants are pure normals. 



In order to prove the former contention, it must be shown that 

 the .F2 aa segregates under the same environment as 402 plants, 

 continuously produce flowers more abnormal than the latter. Further, 

 if contamination is at all common, repeatedly back - crossing normal 

 segregates with the pure abnormal (300 — 309) race, should eventually 

 give one a pure abnormal race, and each generation of back- crosses 

 should show some progress. In making such an experiment, one must 

 first be certain that homozygous lines are being used as material, 

 otherwise, if favorable, one might look upon its results as due to 

 unconscious selection, for if one believes as does Castle (1912) in the 

 creative power of selection, there is no apparent physiological reason 

 why, through self-fertilization and selection, an abnormal race such as 

 300—309 could not be produced from these slightly ab normal -flowered 

 F2 segregates. 



If one adopts the "fixed factor" conception of East (1912) the 

 extra parts of normal flowers on aa and |402 plants are simply and 

 plausibly explained as ontogenetical accidents. For all practical pur- 

 poses, the aa segregates are as normal as (402) individuals. 



Cytology. The maturation phenomena of the anthers were 

 investigated. Cytological irregularities in AA segregates were similar 

 to those of the abnormal grandparent; in Aa anthers, similar to the Fi 



^) The term accident is used throughout this paper in a very specific sense. In 

 using it, 1 intend to convey the idea that is expressed when a very accurate technical 

 device, such as a rotary microtome, misses cutting a section. 



