218 University of California Publications in Botany [Vol. 5 



total germination are nearly identical except in the case of the 

 seed resulting from the cross — i.e., the seed of 1909 H18. 



In general, then, the most significant facts brought out by 

 these last tests have to do with the exceedingly high percentage 

 of germination among the two sets of hybrid seed from one to 

 three years old and the more or less scattering germination of 

 the seed of H20 throughout. In connection with this latter 

 situation the seed of 78/05, as will be seen, shows a remarkably 

 long period during which germination takes place — i.e., over 

 four weeks except in the case of this year's seed and that of 1911. 

 There seems thus to be some difference between the cross and its 

 reciprocal as to the rapidity of germination and the relatively 

 slow germination of 78/05 is seen only in that one of the two 

 hybrids in which this N. T ah acum-var \ety is the female parent. 

 On the other hand the germination of H20 is practically identical, 

 in the length of the period during which germination takes place, 

 with that of 110/05 in general, while the germination of HIS is 

 much more rapid throughout than either 110/05 or 78/05. The 

 significance of these facts is not entirely clear. The increase in 

 amount of total germination in the case of all the hybrid seed 

 as compared with the seed of the parental varieties during the 

 various years is rather striking. In connection with the seed pro- 

 duced by the F^ plants of HI 8 and H20 the increased germina- 

 tion of the hybrid over the parent seed of the same year might 

 be referred to the general stimulating effect of heterozygosis 

 (East and Hayes, 1912). In this connection it may be noted 

 that the plants which bore this Fj hybrid seed showed in a 

 general way an increase in vegetative characters as compared 

 with the parental varieties, but to no such extent as in the F^ 

 hybrids involving N. Tahacum-varieties and .V. sylvestris as 

 parents (cf. Goodspeed. 1913). The fact, however, that these 

 latter hybrids were found to produce very little seed, only part 

 of which would germinate seems to indicate that increase of 

 vegetative characters due to heterozygosis, is not necessarily 

 correlated with the production by such hybrids of abnormally 

 large or even normal amounts of seed or of seed which will show 

 a high percentage of total germination. 



