Florida Agricultural Experiment Station


habit have been found among the later generations of crosses
between Spanish and runner peanuts. Such plants do not have
many-seeded fruits necessarily. Their analogy with Valencia
type is not easily recognized at first because of their usually
long, slender, prostrate branches. Valencia and related vari-
eties except the African forms described by Chevalier have short,
thick, upright branches. The type of hybrid plants with sparse
branching and central stem in the fruit cluster has been named
Valencia plant type. There is no inference as to the number
of seeds per pod.
 The inheritance of Valencia plant type has not been investi-
gated in crosses involving any varieties of the Valencia group
but the data of a cross of Spanish and Jumbo Runner, 1 x 14,
are summarized in Table 11. The record on the F2 is not known
exactly but it is certain that approximately 5 percent were
Valencia type, as was the case in other crosses of Spanish and
runner peanuts. When the F3 generation of cross 1 x 14 was
grown the significance of the sparsely branched plants was
recognized. Seventy-two F3 families were grown to maturity
with an average of 13 plants. The occurrence of Valencia type
plants suggested duplicate gene inheritance with Valencia type
recessive. Theoretical ratios based on that hypothesis are com-
pared with the experimental ratios in Table 12. Because of
the small number of F3 plants in the F2 families it was necessary
to correct the theoretical ratios. This was done by making use
of Warwick's tables (17). In the portion of the table dealing
with 13 progeny tested, it is recorded that 43 percent expected
to breed 15:1 would produce no recessive progeny while the
other 57 percent are expected to produce dominants and re-
cessives in a ratio of 8.085:1. These corrections were made on
the theoretical ratios in the first and second comparisons of
Table 11. No correction for the monohybrid families was neces-
sary. These corrections are not entirely accurate since the
number of F3 progeny per F2 was not constantly 13. A more
accurate correction could probably be made by treating each
F3 family separately and summing the corrections. However,
the error in the method used tends to increase X2 and no false
conclusions are likely.
 Agreement of the data with the hypothesis is quite satis-
factory in three comparisons. It is concluded that Valencia type
is a recessive character controlled by duplicate genes. Peanut
varieties of the Valencia group are probably of the genotype