40 BURSA BURSA-PASTORIS AND BURSA HEEGERI : 



giving- more nearly the aspect of B. bp. heteris than of B. bp. tenuis. I now 

 suspect that these should have been classed with B. bp. tenuis. 



In summing- up these 20 hybrid families belonging- to the third or later 

 generations, it appears that 8 of the 9 possible combinations are included. 

 Two of the seed-plants proved to be extracted B. bp. heteris, 4 were extracted 

 B. bp. tenuis, 2 were extracted B. bp. rhomboidea, 1 was extracted B. bp. 

 simplex; 3 were again like the parent in being di-hybrids which gave rise to 

 all 4 biotypes; 3 were B. bp. heteris X tenuis, 2 were B. bp. heteris X rhom- 

 boidea, 1 was B. bp. rhomboidea X simplex, and the characters of 2 were in 

 doubt. These 2 doubtful families (067 and 0621) were either extracted B. 

 bp. rhomboidea or hybrids between B. bp. rhomboidea and B. bp. simplex, 

 the doubt being caused by the fact that stunted or juvenile specimens of 

 B. bp. rhomboidea may be practically indistinguishable from B. bp. simplex. 

 In a few families the conditions of the culture made the determination of 

 ratios impracticable, but in the majority the approximation to Mendelian 

 expectation is fairly close. Small numbers of B. bp. heteris occurred in 3 

 families where large numbers would have been expected if that elementary 

 form had been a normal component of these hybrid families, since B. bp. 

 heteris dominates both B. bp. tenuis and B. bp. simplex, and to a slight degree 

 B. bp. rhomboidea also. The 8 specimens of B. bp. heteris which seemed 

 out of place among more than 2,600 individuals included in these fami- 

 liesless than one-third of 1 per cent of the whole were probably the 

 result of chance crosses, as all of these occurred in families whose polli- 

 nation had not been guarded. 



040.14 : The second hybrid brought in from nature was collected by C. 

 A. Shull, in Jackson Park, Chicago, in the summer of 1905. This plant 

 had long, acutish lobes, serrated on both margins, and the sinuses extended 

 practically to the rachis. There was a very faint indication of a rounded 

 secondary lobe in some leaves. Seeds were sown December 27, 1905, and 

 produced 99 B. bp. heteris and 26 B. bp. tenuis, or in the ratio 3.8:1. The 

 two components of this hybrid were at first considered distinct biotypes 

 until another family bearing the same characters as the dominant form 

 proved in the third controlled g-eneration to be B. bp. heteris. Subsequent 

 breeding has demonstrated that these hybrids also present the characters 

 of typical B. bp. heteris and typical B. bp. tenuis when grown for several 

 generations under favorable conditions. The peculiarities of the original 

 plant and of the first generation under culture are thus shown to have 

 been fluctuations of these two types. Eig-ht families of the third genera- 

 tion have been studied, with the following results : 



0514.128: This was a specimen of B. bp. tenuis, the recessive form, 

 and as the pollination was carefully guarded, it should have been expected 

 to produce nothing but the parental form among- the offspring. The seeds 



