126 SIRKS, THE COLOURFACTORS OF THE SEEDCOAT IN PHASEOLUS 
combination KLBl (27 from 64), type II contains K1B1 (9 from 64), 
type III KLbl (9) and Klbl (3), type IV k L BI (9) and type V 
k L bl (3), and klbl (1). A comparison of the frequencies obtained 
with those expected according to these suppositions gives a rather 
good agreement: 5 







Type bee nee eat |) av. EVENE 
| | 
1010 Observed frequencies 26 15 | 12 9 | 5 0,627 
Theoretical proportions 27 9 12 9 7 | 64 
Expected numbers. 29 3 13 9 7 Per 

Type II is the only one, that does not wholly accord with the 
expectation; the cultures of F3-families that are grown in this year, 
will decide if these provisional suppositions are well-founded or 
if they are not; the meaning of the more detailed divisions of the 
types into subtypes may be kept also for further research. 
The second of both spontaneous hybrids gave a more limited 
descendance; its examination gave therefore less results than the 
first one. Here also a grouping into five different types could be 
made; type I was divided into two subtypes: one darker la (pl. I 
fig. 5), the other lighter (pl. I fig. 6); type II was a blueviolet striped 
bean with a yellowish-white groundcolour (pl. I fig. 1); type III the 
same but the striping in reddish-violet (pl. I fig. 2), IV the Roties- 
race, and V the Wagenaar-race. Type III may be supposed to be 
identical with the unknown father. The segregation observed in this 
Fy-family did not agree very well with the theoretical expectations, 
as is indicated by the following table: 


Type Deeded elie AU | IV | Vv | Total 
1011 Observed frequencies 3 9 5 3 2.) 22 
Theoretical proportions 27 9 12 9 7 | 64 
Expected numbers. (idees tapes 3 2.117022 
If here again the types with and those without one of the vis- 
ible factors are taken together, the same phenomenon as that in the 
previous cross is met with: with Kl (= the factor for violet striped 
pattern on the yellowish-white groundcolour), types I, II and III, 
