46 
not know whether the Evening Primrose which he investi- 
gated was a pure race or of hybrid origin. = 
To horticulturists a knowledge of Aybrids is of the 
ee importance, a vast number of new and interesting 
forms being continually produced through hybridisation, 
At one time it was thought that hybrids were invariably 
intermediate between the two parental forms, and that they 
were generally sterile, and could therefore not be repro- 
duced by seed. Though this is sometimes the case, 
particularly where the parents are of different species, it 1s 
by no means the rule, and certainly not in the case of 
hybrids between two different varieties. Our exact know- 
ledge of hybrids dates from the careful experiments made 
in the latter half of the nineteenth century by Gregor, 
Mendel, of Brinn. Unfortunately, his important investi- 
gations did not become generally known until the begin- 
ning of the present century. Mendel’s first observations 
were made on various varieties of the garden pea, and he 
obtained the striking result that in crossing two different 
strains the offspring were not of immediate type, but 
generally inherited the characters of one of the parents in 
their entirety. Thus in crossing varieties, of which one 
had round and smooth and the other wrinkled seeds, he 
obtained seeds all of which were round. He therefore 
considered this to be u dominant character, while he called 
wrinkledness recessive. But when the flowers of the hybrid 
plant were subsequently fertilised with their own pollen, 
the seeds they produced were not all round, some of them 
were wrinkled like those of one of the grand parents. On 
carefully counting the number of these recessive types, he 
found that one in four had reverted to the wrinkled tvoe. 
Moreover, he found that of the round seeds some were of 
pure type, and when further cultivated always produced 
round seeds, while others were of hybrid nature and these 
always produced reversions to the parents which had been 
originally crossed. 
He was able, finally, to demonstrate that of the off- 
spring of every hybrid when self-fertilised one quarter 
reverted completely to the female parent, one quarter to 
the male parent, while half of the offspring remained of 
hybrid nature. These resembled the parent, which pos- 
sessed the dominant character but preserved the recessive 
character in a dormant or latent form as was shown by its 
reappearance in a subsequent generation. The accurate 
