242 
GENETICS: CLAUSEN AND GOODSPEED 
even though they do show evident group relationships. On the other 
hand sylvestris apparently is monotypic and is distinctly different from 
the Tahacum group. Now, the study of a large number of varietal 
crosses within the Tahacum group has demonstrated that most char- 
acters are expressed in intermediate degrees in the Fi hybrids and subse- 
quent segregation in further generations indicates that these phenomena, 
although complex, are in accord with normal Mendelian expectation. 
The differences within the Tahacum group, therefore, apparently depend 
upon certain factor differences within a common reaction system. 
When, however, any one of this array of Tahacum varieties is crossed 
with sylvestris, the Fi hybrid very nearly or completely reproduces on a 
larger scale the characters of the particular Tahacum variety concerned 
in the hybrid. This has been found true for a number of Tahacum 
varieties; viz. angustifolia, calycina, macrophylla, macrophylla purpurea, 
*Cavala/ ^Cuba,' and 'Maryland;' descriptions and plates of which based 
on material grown in the University of California Botanical Garden have 
been given elsewhere by Setchell. The completeness of the domination 
of the Tahacum parent in the somatogenesis of these Fi Tahacum-syl- 
vestris hybrids is shown particularly in the crosses involving characters 
which are normally recessive in Tahacum variety hybrids. When caly- 
cina, which produces abnormal, split, 'hose-in-hose' flowers, is crossed 
with Tahacum varieties producing normal flowers, the Fi hybrids produce 
the normal type of flowers with few exceptions. On the other hand in 
marked contrast to the type of behavior in varietal crosses, calycina 
when crossed with sylvestris gives an Fi hybrid which produces only 
calycine flowers. Similarly, the partially parthenocarpic tendency of 
'Cuba,' which is manifested in the retention and normal development of 
many fruits without pollination, although recessive in varietal crosses, 
is so impressed on the ^Cuhsi^ -sylvestris hybrid that all the fruits mature 
normally in spite of the fact that no functional pollen is produced. 
Somatogensis in Fi hybrids of Tahacum with sylvestris seems, therefore, 
to be dominated by the Tahacum system as a unit, so that any particular 
modification of the Tahacum reaction system displays its full possibilities 
in the development of such hybrids. 
Now, if these Fi hybrids of Tahacum with sylvestris represent the 
reaction end-product of two fundamentally dissimilar reaction systems, 
then the relations of these two systems, as manifested by the domina- 
tion of the Tahacum system to nearly or quite the exclusion of the 
sylvestris system, indicate a rather extensive mutual incompatibility of 
the elements of the two systems. This deduction is borne out by the 
fact that the Fi Tahacum-sylvestris hybrids produce only a very few 
