REPRODUCTIVE ORGANS IN THE PRIMULACEE. 95 
hirsuta are given in the table; but I have failed by either form 
pollen to effect the converse unions, 7. e. fertilize the short-styled 
P. hirsuta by either form pollen of the P. Auricula. In the case 
of the long-styled form of P. hirsuta I utterly failed to effect a 
single union, though I tried it homomorphically and heteromor- 
phically by the two-form pollens of P. Auricula, and also con- 
versely by applying the pollen of the long-styled P. hirsuta to 
the two forms of P. Auricula*. These few details then will en- 
able us to consider, in part, the amount of parallelism existing 
between certain of the phenomena of hybridism in normal her- 
maphrodite species, and the hybridism of dimorphie species. In 
the fertilization of hermaphrodite species, several cases occur in 
which the male element of the one, A, for example, fertilizes the 
female element of the other, B, while the male element of B will 
not fertilize the female element of A: so in the hybridism of di- 
morphic species, with the important appanage of an increased 
complication of the conjunctive powers, we find analogous cases. 
Thus, let A Zand As and BI and B s respectively represent the 
long- and short-styled forms of the P. Auricula and P. hirsuta 
given above, then while the male element of A s fertilizes the 
respective female elements of BZ and Bs, that of AJ cannot 
fertilize either of the latter; again, there is a mutual impotence 
between the male elements of B / and B s and the female ele- 
ments of AJ and As. We thus see that a close parallelism exists 
between those remarkable phenomena occasionally observed in 
the reciprocal crossing of species on the one hand, and those ob- 
served in the reciprocal crossing of the two sexual individuals of a 
dimorphic species with those of a distinct dimorphic species on the 
other. In the latter case, as compared with the former, there is 
of course a greater complexity in the functional relations—eight 
crosses being possible between dimorphic species,—and so it is 
* It may be advisable to state the number of flowers fertilized in each of the 
subjoined experiments :—First series, P. Auricula and P. viscosa: I fertilized 
eight flowers of the short-styled P. viscosa by pollen of the short-styled P. Auricula, 
and ten flowers of the long-styled P. Auricula by pollen of the short-styled P. viscosa. 
Second series, P. Auricula and P. hirsuta: of sixteen flowers of the short-styled 
P. hirsuta, one-half were fertilized by pollen of the long-styled, and the other by 
that of the short-styled P. Auricula; again, of the long-styled P. Airsuta, five 
flowers were fertilized by pollen of the long-styled, and five by that of the short- 
styled P. Auricula ; in the converse unions of these, ten long-styled and ten short- 
styled flowers of the P. Auricula were fertilized by pollen of the long-styled 
P. hirsuta, Thus, including the three successful unions of these species given in 
Table III., we find that the above phenomena are the results of eighty-eight 
flowers, in each case carefully fertilized. 
