60 MASS. EXPERIMENT STATION BULLETIN 185. 



Further investigation showed Emerson that the theory of mosaics 

 could not explain the behavior of mottled beans resulting from crosses of 

 non-mottled parents, and he advanced (4) a theory suggested by Shull, 

 which supposed one factor was responsible for mottling in fixed races and 

 a different factor responsible for mottling in heterozygous forms mentioned 

 above, which is visible in heterozygous individuals only. 



In another paper (5) Emerson discusses this theory, and develops 

 another theory suggested by Spillman, which supposes mottling to be 

 due to two factors which may exist separately in the heterozygous mottled 

 forms or coupled in those forms which breed true to the mottled characters. 

 By this theory the facts reported in the present paper may be explained. 



Tschermak carried on numerous investigations of the inheritance of 

 seed coat color in beans along with others with stocks and peas. In his 

 most recent paper (22) he analyzes his results, and is able to account for 

 most of them in a satisfactory fashion by means of simple Mendelian 

 factors. 



Shull (17) advanced the hypothesis of the appearance of the mottling 

 factor only in heterozygous individuals referred to above. 



Kajanus (13) reports investigations of the inheritance of colors and color 

 patterns in garden beans, especially of the behavior of a violet marbled 

 type of mottling due apparently to distinct factors. He reports also on 

 the chemical nature of the pigments involved. 



Jarvis (11) and Tracy (19) have given excellent descriptions and a quite 

 stable nomenclature of common bean varieties, and Freeman (7) describes 

 several types of the Mexican frijoles and teparies, P. acutifolnis var. 

 latifolius. 



Methods. 



At first, commercial seed procured from the trade was used, but begin- 

 ning in 1909 steps were taken to breed pure races, and earlier crosses made 

 with plants grown from commercial seed were, so far as possible, repeated. 

 Evidence indicated that a few of these earlier parents were probably 

 hybrids, and such crosses have been ignored in the consideration of results. 

 In all cases the plants used for crossing have been externally true to type, 

 but as will appear later, it is probable that in some varieties two or more 

 races have been encountered. In such cases no external differences have 

 been observed in the parent plants, though their behavior on crossing 

 revealed different genetic composition. 



In making the crosses the procedure of Emerson has been generally 

 followed; that is, emasculation and pollination have been performed in 

 one operation. This method has given sometimes 30 per cent, or more of 

 successful attempts, and at other times a very low percentage of successes. 

 This is probably due in part to unfavorable environmental conditions, and 

 in part to variations in the procedure, — usually the selection of a female 

 blossom that was not sufficiently mature. In a few cases the resulting 

 plants have been like the female parent, indicating that self-fertilization 



