Stout: The Origin of Dwarf Plants 441 



Bartlett has not discussed these in terms of characters that 

 have been gained, lost, or modified, but does point out that the 

 appearance of these two dwarf forms as sister mutants in the rather 

 large number of individuals "bears a certain degree of resemblance 

 to Mendelian segregation." The case is especially interesting in 

 affording a record of the simultaneous origin of hereditary varia- 

 tions from a single parentage giving dwarf forms of markedly 

 different characteristics. ' 



Furthermore, the facts of variation resulting from hybridization 

 involving dwarf forms indicate a diversity of behavior quite in line 

 with the above-mentioned facts. Mendel considered that the 

 dwarf garden pea differed from the tall in a single character pair 

 and that he obtained neither intermediates nor a greater range of 

 variation in the F2 generation. Darbishire (1911) considers that 

 this is the case and that the dwarf garden pea differs from the tall 

 variety only in respect to two types of internodes, long and short. 

 Punnett (191 1, p. 35) applying the conception of presence and 

 absence very generally to peas states "all peas are dwarf, the tall 

 pea is a dwarf plus a unit factor." This generalization is not borne 

 out by the data. Lock (1904) in progeny of crosses between a tall 

 and a semi-dwarf ("Satisfaction" and native No. 2) found in the F2 

 great variation in habit ranging from very robust to very feeble 

 dwarf. Although he considers that there must be some sort of 

 Mendelian segregation if the internodes are solely considered, he 

 admits that "this cross seems to afford an example of remarkable 

 intensification of both the allelomorphic characters of the same 

 pair, namely tallness and dwarfness, the former in the Fi and both 

 in the F2 and later generations" (p. 414). Keeble and Pellew 

 (1910) in crossing two well-known staple semi-dwarfs varieties of 

 the garden pea obtained in the F2 a w^ide variation, giving both 

 dwarf and tall as well as intermediates. They conclude that length 

 of internode and width of stem are both involved and in their 

 interpretation dwarfs lack both factors, semi-dwarfs lack one and 

 tails possess both. Height in sweet peas is also of complex charac- 

 ter, as breeding studies of dwarf, semi-dwarf and tall varieties have 

 shown. Crosses with bush and "Cupid" give an Fi that is tall and 

 an F2 with variation involving plants that include tall, bush and 



