II4 BOTANICAL GAZETTE ; [FEBRUARY 
produced by using them as pollen parents, the offspring are exactly 
the same as if normal males had been used. In A and B the her- 
maphrodite character is borne by the male germ cells and is fully 
hereditary. In C and D the hermaphrodite character may have been 
purely somatic, in no wise affecting the germ cells, and therefore 
incapable of hereditary transmission; or, in case the factor for 
hermaphroditism is independent of the sex-producing genes, A 
TABLE I 
Pedigree no. Cross Result 
OStign cae %Xself (A) 249:19% 
OEY oie ee 3Xself (A) 92:63 
Oitg fo: 3 Xself (B) I109:95% 
‘Lotal i... 3X self 1439: 1209 
WOU cee. ce 2X8 (A) 582:368 343 
rec cal ee Ge ete 2X¥ (A) 609:408 31d 
GOI ees oX% (A) 5193529 
OBTAQ. got sicy 2X9 (A) 672:33% 
OBIO0 Se 2X8 (B) 532: 508 
OBLI8 os gXY ri 508: 518 
QNI50.25 44... exo (B 592:439 
OBase. ge a5 gX¥(C) . 392:0 3556 
OBTS9. ce eek 2X9 (D) 269:0 +186 
5A and B 3982:3058: 26 
bocce eX ei Ca nd D 652:0 7736 
OS816. 0 ery S{A)X4 21$:29 :116 
and B may have been homozygous and C and D heterozygous with 
respect to the hermaphrodite modifier. 
That males, which are assumed to be heterozygous and to contain 
both the male and female tendencies, should occasionally show the © 
development of characters of both the sexes as the result of some 
accident or environmental influence upon an individual, without in 
any way changing the character of the germ cells produced by that 
individual, is quite conceivable. The occurrence of these two differ- 
ent types of hermaphrodite individuals harmonizes well with many 
experiences met with in other studies in heredity, which have led 
JOHANNSEN (7) to distinguish between genotypes and phenotypes: 
These four hermaphrodite individuals, A, B, C, and D, belong to 4 
single phenotype, but to two genotypes. 
