444 



THE AMERICAN NATURALIST [Vol. LV 



from pure whites, but the facts now known do not suffice 

 to prove this. 



Mutations have apparently occurred in culture in re- 

 spect to both flower color and seed-coat markings. 



This spring (1920) a single dark blue appeared in a cul- 

 ture of j)'mk apricus which had bred true through the three 

 preceding generations. In 1918 a pink arose in a culture 

 of seeds from a wild dark-blue nanus. If the relations 

 between pink and dark blue are the same in the two 

 species one of these cases must be a mutation. Hybridi- 

 zation is exceedingly improbable in the case of the pink 

 apricus. It is not likely that the dark-blue nanus was a 

 hybrid either since it was not collected from a location 

 where this cross would have been likely to occur and 

 only one plant out of a large culture was pink. 



In 1918 plant VIII-27 with dark-blue flowers produced 

 seed of the light color characteristic of light blues. These 

 seeds this year produced both light blues and dark blues. 

 In collecting these seeds it was necessary to read the 

 label on the plant and that on the seed box. It is very 

 unlikely that they faifed to tally with each other or with 

 the color of the flowers still in bloom on the plant. Sev- 

 eral collections were made over a period of two or three 

 weeks and the plant label put in the box at the final 

 collection. Owing to the fact that the seed pods were 

 not opened until after the collections had been finished, 

 no suspicion of anything unusual was entertained until 

 after any sort of check was no longer possible. A mis- 

 take might have been made several times in succession, 

 but this is certainly very improbable. The seeds this year 

 are of the usual type. Three light-blue plants produced 

 light seeds and two dark blues produced no seeds. 



If this is not a case of mistake in records it is very 

 difficult to offer any explanation of it. Since plant VIII 

 was a hybrid and the factor for dark-blue flower color 

 is linked with that for dark seeds, plant VIII-27 could 

 have arisen either through a mutation in the recessive 

 factor for coat color or by a cross-over of its dominant 



