MINUTE STRUCTURE OF PLANT HYBRIDS. 257 



was in full bud in two positions, and had just opened in a third when cut by frost on 

 1st March ; H. ciliatum opened on 23rd April. 



In 1887 I find that the first flowered on 3rd February, the second on 26th February, 

 and the third on 9th April. 



For additional information regarding other hybrid Ehododendrons I would refer to a 

 short article in the Gardeners' Chronicle, vol. ix., 3rd ser., p. 753. 



Many of our hothouse orchid crosses promise, if carefully studied, to give excep- 

 tionally fine results. Hybrids must in not a few cases have been obtained by pollinating 

 an early blossom of an early flowering species with a late blossom of a late flowering 

 species, or at least, if this method has not been adopted, different degrees of temperature 

 have been employed to check or rush forward the flowering period of the plant. Thus 

 Cypripedium Ashbourtonice and C. Harrisianum are examples from a genus. 



Montbretia crocosmcejlora. — During 1890 M. Pottsii opened on 28th July, the 

 hybrid on 20th August, and Tritonia aurea in a cool house on 1st September. 



Lilium Poivellii. — Lily hybrids have been rather rare hitherto. The present one was 

 reared and flowered by C. S. Powell, Esq. of Old Hall, Southborough, to whom I am 

 indebted for a magnificent example. It reached me in full flower at a time when 

 L. Hansoni had shed its parts by a week, and fourteen days previous to the opening of 

 L. dalmaticum. In reply to a query from me, Mr Powell wrote : — " Your observation of 

 the blooming period of the hybrid being intermediate is correct. ' Hansoni ' bloomed 

 before any hybrid expanded, and the other parent, ' dalmaticum,' is now in bloom — ten 

 days after the others (hybrids) were over." 



While the examples now given show the hybrid to be very evenly between its 

 parents, there still are cases which pass to the extremes ; my information, however, 

 about these is in every instance imperfect. But it may be observed here that Dianthus 

 Lindsayi appears to bloom nearly or quite as early as D. alpinus, while D. barbatus, 

 grown from youug plants, opens from a fortnight to three weeks after. The relative 

 age of the specimens here may have something to do with the results. Again, Bryanthus 

 erectus opens shortly after (one to four days) Rhododendron Chamcecistus, according to 

 present statistics, while Menziesia empetriformis, var. Drummondi, opens from fifteen to 

 twenty days later. 



We cannot as yet attempt to formulate definite conclusions, for a large accumulation 

 of statistics from different localities is needed, but the evidence points strongly, and in 

 some cases positively, to the flowering period of hybrids being more or less exactly 

 between that of the parents, while some vary to a greater or less degree towards one 

 or other parent. 



Constitutional Vigour of Hybrids. 



This subject brings us face to face with a very complex and deep-seated chain of 

 phenomena which are the sum-total of the action and reaction of the living protoplasm 

 in relation to its surroundings. The complexity of the subject, and difficulty of judging 



