290 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



except a lessening of the purple colour on the top, I cannot see 

 any change that "Model White " has effected on its companion. 

 Both, of course, are white-fleshed. " Early Ked Milan " seems to 

 have changed the character of " Model White " to a purple-topped 

 variety, and " Model White " to have given its partner a greater 

 depth of flesh. Most of the other varieties were more or less 

 crossed, thus showing the dangers that must be avoided by a 

 careful system of isolation in order to secure a true strain. 



Speaking on this matter leads me on to say that pollen of one 

 variety of a plant may be more active than that of another 

 variety. In the year 1895 some bulbs of " Cranston's Excelsior " 

 onion were planted alongside specimens of " Dobbies' Selected 

 Red." The idea in view was to determine how far they would 

 cross with each other, and if possible produce a new onion. The 

 seed saved from " Excelsior " produced plants and bulbs to all 

 appearance — with only two exceptions — the same as the parent. 

 The two exceptions were bulbs of a light brown colour. On the 

 other hand the seed from the red variety gave just the opposite 

 result. The bulbs resembling the seed parent were very few 

 indeed, the majority being of a reddish brown colour and much 

 deeper in form than the red variety that produced them. The 

 pollen of the " Excelsior " variety seems to have almost entirely 

 overpowered the other. Mr. Martin Smith, the well-known 

 carnation specialist, finds considerable variation in the powers of- 

 the male parent to reproduce itself. With some varieties, such 

 as the yellows, the pollen from plants of another colour seems all- 

 powerful, but, on the other hand, the pollen of yellow varieties 

 does not seem to have any appreciable effect on varieties of 

 different colour. (See the " Royal Horticultural Society's 

 Journal " of August, 1896, p. xlii.) 



Where seed-growing is carried on to any great extent it will 

 be reasonable to expect that certain variations will take place in 

 the habit, colour, or other qualities of the plant or flower. A 

 sharp look-out for these "breaks" must bo kept, and any 

 tendency to develop in a desirable direction must be encouraged. 

 Quite new forms, types, or colours may be permanently fixed in 

 many, or even in most, cases, although it may in some instances 

 take years to accomplish. " Antirrhinum Yellow Queen," which 

 this year received the x x x after trial at Chiswick, was only 

 secured after several years of isolation, inbreeding, and selection ; 



